JP6389129B2 - Intercom system for housing complex - Google Patents

Description

  The present invention relates to a collective housing intercom system capable of detecting an intruder.

  Conventionally, in order to increase security, an automatic door is generally installed at an entrance of an apartment house such as an apartment. In addition, there is a collective housing intercom system in which a camera is provided in a collective entrance machine installed at the collective entrance, and a visitor image captured by the camera is displayed on a display unit of the collective entrance machine (see Patent Document 1).

JP 2011-205570 A

  However, in the system of Patent Document 1, an intruder who has entered through an automatic door of a collective entrance or a intruder that has entered through an open back door is identified as a resident, a visitor, and their accompanying persons. It was difficult. For this reason, the conventional apartment house intercom system has room for improving security.

  Accordingly, an object of the present invention is to provide an intercom system for collective housing that can improve security against an intruder using a companion or the like.

In order to achieve the above object, an intercom system for an apartment house according to the present invention includes:
A collective entrance machine having a collective entrance camera and an input unit for inputting room number information;
A human presence sensor that senses a person, a slave camera that is activated based on the output of the human sensor, and an entrance slave that is installed at the entrance of each room;
When the room number information is input to the input unit, the entrance imaging data captured by the collective entrance camera and the room number information are transmitted from the collective entrance machine, and the human sensor detects a person A controller that transmits the slave image data captured by the slave camera to the entrance slave unit;
With
An extraction unit for extracting characteristics of a person included in the entrance imaging data;
A predicted route determination unit that determines a predicted route of the person based on the room number information;
A recording unit for recording the feature and the predicted route in association with the person;
When the child image data transmitted from the entrance child device of each room includes the characteristics of a person imaged by the collective entrance camera, the image was captured by the child camera based on the child device image data. A flow line estimation unit for estimating a flow line of a person,
A determination unit configured to determine whether the person imaged by the slave camera is an intruder by comparing the predicted path with the flow line;

  According to this configuration, by comparing the predicted path and the flow line, it is determined whether the person imaged by the slave camera is traveling on the same path as the predicted path or traveling on a deviated path. be able to. For this reason, it is possible to determine that a person who deviates from the predicted route is an intruder who has entered the apartment house with a visitor or a resident. As a result, it is possible to find an intruder who has entered the apartment house, and the security against the intruder using the companion can be improved.

Further, in the intercom system for collective housing of the present invention,
When the entrance door of the living room is opened and closed, opening / closing information indicating that the entrance door has been opened and closed is transmitted from the living room to the controller,
The extraction unit extracts a feature as a person who is out of the person imaged by the child camera in the living room that transmitted the opening / closing information,
The recording unit records the opening / closing information and the characteristics of the outing person in association with the outing person,
The determination unit may determine that the person imaged by the slave device camera is not an intruder when the person imaged by the slave device camera is the outing person.

  According to this configuration, for example, a resident who tries to visit another room in the apartment house is determined as a person who has gone out and is not determined as an intruder. In addition, a visitor who returns home is also determined as a person who has gone out and is not determined as an intruder. Thereby, the precision which judges an intruder can be improved.

Further, in the intercom system for collective housing of the present invention,
If the characteristics of the person imaged by the collective entrance camera are not included in the slave device imaging data, the determination unit may determine that the person imaged by the slave camera is an intruder.

  According to this configuration, it is possible to determine that a person who has not been captured by the collective entrance camera is an intruder. For this reason, it is possible to improve the security against an intruder who tries to enter the apartment house by using co-existing etc. while escaping the imaging range of the collective entrance camera.

Further, in the intercom system for collective housing of the present invention,
When the characteristics of the person imaged by the collective entrance camera is not included in the slave device imaging data,
The extraction unit extracts a feature as an unknown person included in the handset imaging data,
The predicted route determination unit determines the predicted route of the unknown person based on the predicted route of the person imaged by the collective entrance camera,
The flow line estimation unit estimates the flow line of the unknown person based on the handset imaging data,
The determination unit may determine whether or not the unknown person is an intruder by comparing the unknown person's predicted route with the flow line of the unknown person.

  According to this configuration, for a person who is outside the imaging range of the collective entrance camera, the person is a visitor's companion or the like, or is an intruder who has entered the apartment house using a companion It can be determined whether or not.

Further, in the intercom system for collective housing of the present invention,
The recording unit may record at least one of the handset imaging data including the person determined to be an intruder and the entrance imaging data.

  According to this configuration, the state of the intruder can be confirmed using the recorded data even after the intruder has left the apartment house. Also, the recorded data can be used as evidence that a suspicious person has entered.

Further, in the intercom system for collective housing of the present invention,
The controller may include a notification unit that notifies a warning sound when the determination unit determines that the person imaged by the slave camera is an intruder.

  According to this configuration, it is possible to suppress the suspicious behavior of the intruder by alerting the intruder in the apartment house with a warning sound, and to call attention to the residents and visitors Can do.

Further, in the intercom system for collective housing of the present invention,
The controller may include a notifying unit for notifying a result of the determination to the management room base unit of the apartment house when the determining unit determines that the person imaged by the slave camera is an intruder.

  According to this configuration, it is possible to notify the manager that an intruder has entered the apartment house and call attention.

Further, in the intercom system for collective housing of the present invention,
When the determination unit determines that the person imaged by the slave camera is an intruder, the controller transmits the slave unit image data including the person imaged by the slave camera to the management room master unit You may send to.

  According to this configuration, the administrator can confirm the intruder in real time using the handset imaging data.

Further, in the intercom system for collective housing of the present invention,
It has a surveillance camera installed in the apartment house,
The determination unit may determine whether the person imaged by the monitoring camera is an intruder.

  According to this configuration, it is possible to quickly and accurately determine whether or not an intruder is using the video information of the surveillance camera.

Further, in the intercom system for collective housing of the present invention,
At least one of the extraction unit, the predicted route determination unit, the recording unit, the flow line estimation unit, and the determination unit includes the controller, a server device installed in an apartment house, and a server device on an external cloud. You may arrange | position in either.

  According to this configuration, it is possible to distribute the processing load of each unit for determining an intruder, and it is possible to maintain a stable operation of the entire system.

Further, in the intercom system for collective housing of the present invention,
The determination unit, when the flow line of the person imaged by the slave camera deviates from the predicted path, when the person imaged by the slave camera returns to the predicted path within a predetermined time, The person may be determined not to be an intruder.

  According to this configuration, the comparison reference between the flow line and the predicted route can be relaxed within a predetermined range, and the possibility that a visitor is erroneously determined as an intruder can be reduced.

  According to the intercom system for an apartment house of the present invention, it is possible to improve the security against an intruder using a shared person or the like.

It is a block diagram of the intercom system for apartment houses which concerns on embodiment of this invention. It is a functional block diagram of the collective entrance machine in the intercom system for collective housing. It is a functional block diagram of the entrance cordless handset in the intercom system for apartment houses. It is a functional block diagram of the controller in the intercom system for apartment houses. It is a schematic diagram which shows the inside of the apartment for demonstrating operation | movement of the intercom system for apartment houses. It is a flowchart for demonstrating operation | movement of the intercom system for apartment houses. It is a flowchart for demonstrating operation | movement of the intercom system for apartment houses. It is a flowchart for demonstrating operation | movement of the intercom system for apartment houses. (A) is a figure which shows an example of the visitor information recorded on the data table, (b) is a figure which shows an example of resident information, (c) is a figure which shows an example of go-out information. .

  Hereinafter, an example of this embodiment will be described in detail with reference to the drawings.

  As shown in FIG. 1, the intercom system 1 for a collective housing includes a collective entrance machine 10, a collective entrance door 40, a monitoring camera 70, a controller 80, a base unit 20, an entrance slave unit 30, and an entrance. A door 50 and a management room master unit 60 are provided.

  The collective entrance machine 10 is installed in a collective entrance that is a shared space of a collective housing. The collective entrance machine 10 has a function of calling a resident or a manager, a function of capturing a visitor or the like with a camera, and the like. The collective entrance machine 10 is connected to the controller 80 via a signal line L1.

  The collective entrance door 40 is provided at the collective entrance of the collective housing, and is formed of an automatic door having a function of restricting access to the living area. The collective entrance door 40 is connected to the controller 80 via a signal line L4. A monitoring camera 70 is installed at the collective entrance. The surveillance camera 70 is connected to the controller 80 via a signal line L6. The surveillance camera 70 is installed not only at the entrance hall but also in front of an elevator on each floor (in this example, the first floor to the second floor).

  The controller 80 has a function of controlling the collective entrance machine 10, the living room master machine 20, the collective entrance door 40, the management room master machine 60, the monitoring camera 70, and the like. The controller 80 is connected to a server device 81 provided in the apartment house and a server device 82 on an external cloud.

  The room master unit 20 is installed in each room of the apartment house. The room master 20 is connected to the controller 80 via the signal line L2.

  The entrance cordless handset 30 is installed at the entrance of each room of the apartment house. The entrance cordless handset 30 has a function of capturing a visitor or the like with a camera. The entrance cordless handset 30 is connected to the living room base phone 20 via the signal line L3. The entrance door 50 is a door provided at the entrance of each living room. The entrance door 50 includes a detection sensor that detects that the entrance door has been opened and closed. The detection sensor of the entrance door 50 is connected to the room master 20 and sends the detected opening / closing information of the door to the room master 20.

  The management room base unit 60 is installed in the management room of the apartment house. The management room master unit 60 is connected to the control unit 80 via the signal line L5.

  As shown in FIG. 2, the collective entrance machine 10 includes a collective entrance camera 101, an input unit 102, a call button 103, a display unit 104, a call unit 105, a collective entrance CPU 106, and a collective entrance interface ( (Hereinafter referred to as a collective entrance machine I / F) 107.

  The collective entrance camera 101 is a camera for imaging visitors and the like. As the collective entrance camera 101, a camera capable of imaging a range of a predetermined angle centered on a visitor who operates the collective entrance machine 10 is used. The input unit 102 is operated for a visitor or resident to input room number information. The room number information includes a room number, a card key ID, and the like. The room number information can be input by operating a button on the operation panel including a numeric keypad, operating a button on the touch panel, touching a card key, or the like. The call button 103 is a button used by a visitor to call a visited place, and is operated after the room number information is input.

  The display unit 104 is, for example, a liquid crystal screen on which a room number input by the input unit 102, a visitor captured by the collective entrance camera 101, and the like are displayed. The call unit 105 includes a microphone and a speaker for making a call with a resident or a manager. The collective entrance machine CPU 106 controls the operation of each part constituting the collective entrance machine 10. The collective entrance machine I / F 107 forms a bidirectional signal transmission path between the collective entrance machine CPU 106 and the signal line L1.

  As shown in FIG. 3, the entrance slave unit 30 includes a human sensor 301, a slave camera 302, a call button 303, a call unit 304, an entrance slave unit CPU 305, and an entrance slave unit I / F 306. ing.

  The human sensor 301 is a sensor that detects a person in front of the entrance cordless handset 30. The human sensor 301 is connected to the child camera 302 and outputs a detection signal notifying that a person has been detected to the child camera 302. The handset camera 302 is a camera for picking up images of visitors and persons passing in front of the entrance handset 30. The slave camera 302 is activated based on the detection signal output from the human sensor 301. The slave camera 302 can capture a range of a predetermined angle in the same manner as the collective entrance camera 101. The handset image data captured by handset camera 302 is transmitted to living room base unit 20 via signal line L3.

  The call button 303 is a button that a visitor presses to call a resident. The call unit 304 includes a microphone and a speaker for making a call with a resident. The entrance slave device CPU 305 controls the operation of each unit constituting the entrance slave device 30. The entrance child device I / F 306 forms a bidirectional signal transmission path between the entrance child device CPU 305 and the signal line L3.

  As shown in FIG. 4, the controller 80 includes an extraction unit 801, a predicted route determination unit 802, a flow line estimation unit 803, a determination unit 804, a recording unit 805, a notification unit 806, and a notification unit 807. , A controller CPU 808 and a controller I / F 809 are provided. The entrance image data captured by the collective entrance camera 101 is transmitted from the collective entrance machine 10 to the controller 80 via the signal line L1. Further, the room number information input to the input unit 102 of the collective entrance machine 10 is transmitted from the collective entrance machine 10 to the controller 80 via the signal line L1. In addition, slave device image data captured by the slave device camera 302 is transmitted from the living room master device 20 to the controller 80 via the signal line L2. Further, an unlocking permission signal for unlocking the collective entrance door 40 is transmitted from the living room master 20 via the signal line L2 to the controller 80. Further, a command signal from the management room parent device 60 is transmitted to the controller 80 via the signal line L5. In addition, image data captured by the monitoring camera 70 is transmitted to the controller 80 via the signal line L6.

  The extraction unit 801 analyzes video information such as entrance image data and child device image data, and extracts features such as a face, body, and clothes of the person being imaged. The entrance imaging data and the slave imaging data include not only moving image data but also still image data. Based on the room number information input to the input unit 102, the predicted route determination unit 802 determines a predicted route from the collective entrance predicted to pass the person included in the entrance imaging data to the target room. The predicted route is not limited to the shortest route from the collective entrance to the living room, but can be defined to include a route including a route passing through a shared facility (a toilet, a vending machine corner, etc.). The flow line estimation unit 803 estimates (calculates) a flow line that is a path actually taken by a person imaged by the child device camera 302 based on the child device image data captured by the plurality of child device cameras 302. .

  The determination unit 804 determines whether or not the person imaged by the handset camera 302 is an intruder (suspicious person) by comparing features, predicted routes, flow lines, and the like. In the recording unit 805, visitor information 811, resident information 812, outsider information 813, entrance imaging data 814, handset imaging data 815, and the like are recorded in association with the imaged person. When the determination unit 804 determines that an intruder has entered the apartment house, the notification unit 806 notifies the user of the fact by outputting a warning sound, for example. When the determination unit 804 determines that an intruder has entered the apartment house, the notification unit 807 notifies the management room master unit 60 of the determination result via the signal line L5. The controller CPU 808 controls the operation of each part constituting the controller 80. The controller I / F 809 forms a bidirectional signal transmission path between the controller CPU 808 and the signal lines L1, L2, L4, L5, and L6.

  Next, an operation for determining whether or not a person is an intruder will be described with reference to FIGS. In this example, the apartment house is a condominium having a configuration as shown in FIG. In addition, the following persons are introduced and explained. In the following description, an “intruder (suspicious person)” means a malicious person who uses a shared person to enter a condominium and, for example, tries to be a thief in some room. “Unknown person” means a person who is unclear whether it is a good-willed person or a malicious person.

<Character>
・ Resident M (visitor in Room 204, not shown in FIG. 5)
Visitor A (a person who has operated the collective entrance machine 10 at the time of visit and was captured by the collective entrance camera 101)
・ Visitor B (The person who does not operate the collective entrance machine 10 at the time of visit, but who visited with the visitor A and was captured by the collective entrance camera 101)
・ Visitor C (person who enters the apartment with a visitor or a resident and is captured by the collective entrance camera 101)
・ Visitor D (unknown person who visited with visitors A and B but was not captured by the entrance camera 101)
・ Visitor E (a person who enters the apartment with a visitor or a resident and is not captured by the collective entrance camera 101)
・ Resident N (person who visits a resident in a separate room and moves inside the apartment)

<Character of Pattern 1>
・ Visitor A
・ Visitor B
・ Visitor C
・ Resident M
As shown in FIG. 6, a visitor A inputs room number information (room 204) from the input unit 102 of the collective entrance 10 and calls a resident M (step S101).
When calling is performed by operating the collective entrance machine 10, the collective entrance camera 101 of the collective entrance machine 10 is activated to capture an image of a visitor within the imaging range of the collective entrance camera 101 (step S102). In this example, visitors A, B, and C are imaged in the imaging range S (see FIG. 5) of the collective entrance camera 101. The collective entrance machine 10 may be provided with a human sensor, and when the human sensor detects a visitor, the collective entrance camera 101 is activated to capture a video of the visitor.

  The input room number information (room 204) and the entrance imaging data imaged by the collective entrance camera 101 are transmitted to the controller 80 via the signal line L1. The extraction unit 801 of the controller 80 analyzes the entrance imaging data and extracts the characteristics of each of the three visitors A, B, and C that are being imaged (step S103). The extracted features (for example, features A, B, and C) are associated with room number information (for example, visitor A: feature A: room 204, visitor B: feature B: room 204, visitor C: feature C). : Room 204), it is recorded as visitor information in the data table of the recording unit 805 (see step S105, FIG. 9A).

  Subsequently, the predicted route determination unit 802 determines the predicted routes R1 of the visitors A, B, and C captured together by the collective entrance camera 101 based on the input room number information, passage map information in the apartment, and the like. Determine (step S104). Specifically, the predicted route determination unit 802 indicates that the visitors A, B, and C “go through the entrance door 40 and ride on the first floor elevator 61 or go up the stairs to the second floor and go from the room 201 to the room 203. The predicted route R1 is determined as “passing through the front of the room sequentially and heading to the room 204”. The determined predicted route R1 is recorded as visitor information in the data table of the recording unit 805 (see FIG. 9A).

  Subsequently, it is determined whether or not the lock of the collective entrance door 40 is unlocked by the resident M. If the unlocking is not permitted, the process returns to step S101 (step S106: No). On the other hand, when the unlocking is permitted, the controller 80 unlocks the lock of the collective entrance door 40 and then waits for the slave unit image data to be transmitted from the entrance unit 30 (step S106: Yes).

  Visitors A and B enter the living area when the collective entrance door 40 is opened, and take the elevator or go up the stairs to the second floor and go to room 204 (see FIG. 5). On the way, when the visitors A and B first pass through the collective entrance door 40, the surveillance camera 70 in front of the collective entrance door 40 images the visitors A and B. When the visitors A and B get on the elevator 61 on the first floor, the monitoring camera 70 in front of the elevator 61 is displayed. When the visitors A and B get off the elevator 61, the monitoring camera 70 in front of the elevator 61 on the second floor is displayed. , B are imaged. Furthermore, when the visitors A and B pass in front of the entrance cordless device 30 in the rooms 201 to 203, the human sensor 301 of each entrance cordless handset 30 detects the visitors A and B, and each cordless handset camera 302 visits. The persons A and B are imaged (step S107). The child device image data obtained by capturing the images of the visitors A and B are sequentially transmitted to the controller 80 by each of the imaged entrance child devices 30 via the living room parent device 20. In this example, each surveillance camera 70 is always activated to capture an image.

  Subsequently, the extraction unit 801 analyzes the child device image data sent from each entrance child device 30 in the rooms 201 to 203, and extracts the characteristics of each person being imaged (step S108). The determination unit 804 determines whether the extracted feature of each person matches one of the feature A, feature B, or feature C recorded in the data table (step S109).

  In this example, it is determined that the feature of the visitor A captured by each slave device camera 302 matches the feature A stored in the data table. Further, it is determined that the feature of the visitor B matches the feature B. Accordingly, the determination unit 804 is the same person as the visitor A in which the person having the feature A is imaged by the collective entrance camera 101, and the same as the visitor B in which the person having the feature B is imaged by the collective entrance camera 101. Judged to be a person.

  When it is determined that the features match, the flow line estimation unit 803 acquires the imaging time of each of the slave device imaging data sequentially sent from each entrance slave device 30. The flow line estimation unit 803 analyzes the direction in which the person has moved based on each imaging time, and passes through the flow lines of the captured visitors A and B (for example, the direction of the room 202 through the front of the room 201). Is advanced (in the direction of room 203 after passing in front of room 202) (step S110). The determination unit 804 determines whether the estimated flow line matches the predicted route R1 recorded in the data table (step S111).

  In this case, the determination unit 804 determines that the flow lines of the visitors A and B coincide with the midway route “sequentially passing through room 201 to room 203” which is a part of the predicted route R1. The determination unit 804 determines that the visitors A and B are captured by the collective entrance camera 101 when the resident M is called in the collective entrance 10 and are traveling along the predicted route R1. , B is not an intruder (step S112).

  Even if the traffic lines of the visitors A and B captured by the slave camera 302 deviate from the predicted route R1, the visitors A and B enter the predicted route R1 within a predetermined time (for example, 30 seconds). When returning, it may be determined that a detour has been made in the middle, and it may be determined whether or not it matches the predicted route R1 by excluding the deviated flow line. In addition, in the determination of the intruder, data captured by each monitoring camera 70 may be used in addition to the entrance imaging data and the slave device imaging data.

  On the other hand, when the collective entrance door 40 is opened, the visitor C enters the apartment together with the visitors A and B. Since visitor C is not the intended person to visit resident M, visitor A and B take different actions. In that case, when the visitor C passes the collective entrance door 40 and gets on and off the elevator 61, each surveillance camera 70 images the visitor C. Further, when the visitor C passes in front of the entrance cordless handset 30 in any room, each human sensor 301 senses the visitor C and each handset camera 302 images the visitor C. For example, when the visitor C passes in front of the 301-302 rooms, the subunit | mobile_unit camera 302 of the entrance subunit | mobile_unit 30 of each living room images the visitor C (step S107). The slave unit imaging data obtained by imaging the visitor C is sequentially sent to the controller 80.

  As in the case of the visitors A and B described above, the characteristics of the visitor C are first compared (step S109). The visitor C is imaged by the collective entrance camera 101 together with the visitors A and B. Therefore, it has the feature C recorded in the data table. Therefore, the determination unit 804 determines whether the flow line of the visitor C is coincident with the predicted route R1 (step S111).

  In this case, the determination unit 804 determines that the flow line of the visitor C (route passing in front of room 301 to 302) does not coincide with the predicted route R1 at all. The determination unit 804 captures the image of the visitor C by the collective entrance camera 101 when the resident M is called in the collective entrance machine 10 but is acting as a visitor who is acting on a route deviating from the predicted route R1. The visitor C is judged to be an intruder (step S113).

  The notification unit 807 notifies the management room base unit 60 and the like together with the characteristics of the intruder (feature C of the visitor C) and the like that the intruder has been confirmed. The notification unit 806 outputs an alarm sound to notify the apartment that there is an intruder in the apartment. In addition, the controller 80 transmits the slave unit imaging data including the intruder C captured by the slave unit camera 302 to the management room master unit 60.

In addition, when the person who is going to enter the apartment is, for example, a resident M who has returned home from the outside, the following operation is performed.
The resident M inputs room number information using a key such as a card key at the collective entrance 10. The determination unit 804 determines that the person who is about to enter the apartment is the resident M in the 204th room because the room number information (No. 204) is input using the key. The controller 80 immediately unlocks the collective entrance door 40 for the resident M. The feature (for example, feature M) of the resident M is recorded in advance as resident information in the data table of the recording unit 805 (see FIG. 9B). If it is determined in step S109 that the feature of the person captured by the handset camera 302 matches the feature M recorded in the data table, the determination unit 804 does not determine the flow line. It is determined that the person is a resident M and not an intruder.

  According to the collective housing intercom system 1 of the present embodiment described above, the characteristics of the person imaged by the slave camera 302 are imaged by the collective entrance camera 101 when determining whether or not it is an intruder. The characteristics of the visitors A, B, and C are compared, and the flow line of the person captured by the slave camera 302 is compared with the predicted route R1 of the visitors A, B, and C. For this reason, it is determined whether the person imaged by the handset camera 302 is the visitor A, B, or C, and the visitor A, B, or C is traveling on the same route as the predicted route R1 or deviates. It can be determined whether the route is in progress. Thereby, it can be determined whether the visitor imaged with the collective entrance camera 101 is an intruder. Therefore, it becomes possible to find the intruder who entered the apartment, and the security against the intruder using the companion can be improved. Moreover, by performing the same comparison as described above, it is possible to determine that the resident M who has returned home from the outside is not an intruder, and can be prevented from being erroneously determined as an intruder.

  Further, when estimating the flow line of the person imaged by the handset camera 302, even if the person detours on the way to the home and proceeds along the path deviating from the predicted path, the predicted path is set within a predetermined time. If it returns, it will not be judged that it deviated from the prediction route, and the person will not be judged as an intruder. For this reason, even if a person takes an action (an approach) that is difficult to predict on the way to the visited home (room 204), the action can be excluded to determine whether the person is an intruder.

  When the presence of an intruder is confirmed, at least one of the handset imaging data and the entrance imaging data including the intruder is recorded in the recording unit 805. For this reason, even after the intruder leaves the condominium, the state of the intruder can be confirmed using the recorded data. The recorded data can be used afterwards as evidence that an intruder has invaded. In addition, since the video information of the surveillance camera 70 that is always activated can be used when making an intruder determination, it is possible to determine whether or not the intruder is quicker and more accurate.

  Further, when the presence of an intruder is confirmed, a warning sound is notified in the apartment, so that the suspicious behavior of the intruder can be suppressed by the warning sound. In addition, attention can be drawn to residents and visitors. In addition, since the management room base unit 60 is notified that the presence of an intruder has been confirmed, the manager's attention can be alerted. Furthermore, since the handset imaging data including the intruder is transmitted to the management room base unit 60, the administrator can confirm the intruder in real time using the handset imaging data.

<Character of Pattern 2>
・ Resident N (for example, when Resident N in Room 304 visits Room 301)
・ Visitors A, B, and D (when leaving the room 204 and returning home)
As shown in FIG. 7, for example, when the resident N in the 304th room opens the entrance door 50, opening / closing information of the entrance door notifying that is sent from the room master 20 in the 304th room to the controller 80 (step S201).

  When the resident N goes out of the room 304, first, the human sensor of the entrance handset 30 in the room 304 senses the resident N, and the handset camera 302 images the resident N (step S202). The handset image data in which the resident N is imaged is transmitted from the entrance handset 30 to the control unit 80 via the room base unit 20.

  The extraction unit 801 of the control device 80 analyzes the handset image data sent from the entrance handset 30 in the room 304, and extracts the characteristics of the person being imaged (step S203). The extracted feature of the person (for example, feature N) is associated with the opening / closing information of the entrance door and the room number (for example, outing person: feature N: Room 304: going out), and is stored in the data table of the recording unit 805 as outsider information. Record (see step S204, FIG. 9C).

  Subsequently, when the resident N passes in front of the entrance cordless handset 30 in the rooms 303 to 302, each human sensor senses the resident N and each handset camera 302 images the resident N (step S205). The slave unit imaging data in which the occupant N is captured is sequentially transmitted from the captured entrance subunits 30 to the controller 80.

  The extraction unit 801 analyzes the child device image data transmitted from each entrance child device 30, and extracts the characteristics of the person being imaged (step S206). The determination unit 804 determines whether the extracted feature of the person matches the feature N recorded in the data table (step S207). In this example, it is determined that the feature of the resident N captured by the handset camera 302 in the 303 and 302 rooms matches the feature N recorded in the data table (step S207: Yes).

  Accordingly, the determination unit 804 determines that the person having the feature N captured by the handset camera 302 in the 303 and 302 rooms is the same person as the person captured by the handset camera 302 in the 304 room. Then, the determination unit 804 determines that the person imaged by the handset camera 302 in the room 304 is a person who has gone out of the room 304 and is not an intruder (step S208). On the other hand, when the person imaged by the handset camera 302 in the rooms 303 and 302 does not have the feature N recorded in the data table (step S207: No), the determination unit 804 selects the person. It is determined that the person is an intruder (step S209). Thereafter, for example, also when the resident N is imaged by the slave camera 302 of the other entrance slave unit 30, the determination unit 804 similarly determines that the resident N is not an intruder based on the outsider information.

  Similarly, when the visitors A, B, and D leave the resident M's home (room 204) and return home, the entrance door opening / closing information and the characteristics of the person captured by the handset camera 302 are also displayed. Based on this, the determination unit 804 determines that the visitors A, B, and D are outing persons and not intruders. The characteristics (for example, characteristics A, B, D) of the person extracted by the extraction unit 801 are associated with the opening / closing information of the entrance door and the room number (for example, outing person: characteristic A: Room 204: outing, outing person: The data is recorded in the data table of feature B: room 204: going out, person going out: feature D: room 204: going out (see FIG. 9C).

  When the manager goes out of the management room, it is determined that the person is not an intruder in the same manner as in the above pattern 2.

  According to the above operation, when determining whether or not the person is an intruder, the opening / closing information of the entrance door 50 and the handset image data transmitted from the entrance handset 30 of the room that outputs the opening / closing information are used. ing. Thus, for example, in a condominium, the resident N in the room 304 who wants to visit another room (room 301) is determined to be an outing person and not an intruder. In addition, visitors A, B, and D who return home from the destination (room 204) are also determined to be outing persons and are not determined to be intruders. Accordingly, it is possible to suppress erroneous determination of a person who is not an intruder as an intruder.

<Character of Pattern 3>
・ Visitor A
・ Visitor E (unknown person)
・ Resident M
In FIG. 8, each step from step S301 to step S313 is processed in the same manner as each step from step S101 to step S113 in FIG. Thereby, it is determined that visitor A is not an intruder in step S312.

  On the other hand, since the visitor E is a person who has entered the apartment with the visitor A, the visitor E does not have the purpose of visiting the resident M and takes a different action from the visitor A. When the visitor E passes in front of any room, the handset camera 302 of the entrance handset 30 images the visitor E. The slave unit imaging data obtained by imaging the visitor E is sent to the controller 80 (step S307).

  The determination unit 804 compares the characteristics of the visitor E extracted from the handset imaging data with the characteristics of the person (visitor A) extracted from the entrance imaging data (step S309). In this case, the visitor E is not imaged because he was outside the imaging range S of the collective entrance camera 101 (see FIG. 5), and does not have the characteristics of the person imaged in the entrance imaging data (step S309). : No). Therefore, the determination unit 804 determines that a person (visitor E) that has not been captured by the collective entrance camera 101 is an intruder (step S313).

  According to the above operation, if the person captured by the handset camera 302 is a person not captured by the collective entrance camera 101, the person is determined to be an intruder. For this reason, it is possible to find an intruder (mysterious visitor E) who tries to enter the apartment with the visitor A while escaping the imaging range of the collective entrance camera 101 so as not to be captured by the collective entrance camera 101. Become. Thereby, the security with respect to an intruder can further be improved.

<Character of Pattern 4>
・ Visitor A
・ Visitor D (unknown person)
・ Visitor E (unknown person)
・ Resident M
As shown in FIG. 6, the visitor A is processed in each step in the same manner as the visitor A of the pattern 1 and is determined not to be an intruder in step S112.

  Next, the visitor D visits the resident M together with the visitor A, and goes to the resident M's room (room 204) together with the visitor A. On the way, when visitors A and D pass in front of entrance cordless handset 30 in rooms 201 to 203, each handset camera 302 images visitor A and D, and handset image data is sequentially sent to controller 80. (Step S107).

  The extraction unit 801 extracts the characteristics of the person being imaged in the same manner as described above (step S108). The determination unit 804 determines whether the extracted person (visitor D) has the same characteristics as the person A recorded in the data table (the characteristics of the visitor A captured by the collective entrance camera 101) ( Step S109). In this case, since the extracted person (visitor D) is outside the imaging range S of the collective entrance camera 101 (see FIG. 5), it is not captured by the collective entrance camera 101. For this reason, it is determined that the feature of the extracted person (visitor D) does not match the feature A. Accordingly, the determination unit 804 determines that the person who does not have the feature A (visitor D) is the unknown person X. The extracted characteristic of the unknown person X (for example, characteristic D) is recorded as visitor information in the data table in association with, for example, “unknown person X: characteristic D” (see step S121, FIG. 9A).

  Subsequently, the flow line estimation unit 803 obtains the time of capturing each of the handset image data in the same manner as described above, and estimates (calculates) the flow line of the captured unknown person X (step S122). The determination unit 804 determines whether the flow line of the unknown person X matches the predicted route R1 recorded in the data table (the predicted route of the visitor A captured by the collective entrance camera 101) (step S123). In this case, the unknown person X is a companion of the visitor A, and the flow line thereof is the same as the flow line of the visitor A. The determination unit 804 determines that the flow line of the unknown person X matches the predicted route R1. The determination unit 804 determines that the unknown person X is a visitor (friend, family, etc.) accompanying the visitor A and is not an intruder (step S124). Then, the visitor information in the database is updated to, for example, “accompanying visitor D: feature D: room 204: R1” (see FIG. 9A).

  On the other hand, since the visitor E is a person who has entered the apartment with the visitor A, the visitor E takes a different action from the visitor A. At this time, when the visitor E passes in front of the entrance cordless handset 30 in one of the rooms, the handset camera 302 of the entrance cordless handset 30 images the visitor E (step S107). The slave device image data obtained by imaging the visitor E is sent to the controller 80.

  Since the visitor E was outside the imaging range S of the collective entrance camera 101 (refer to FIG. 5) like the visitor D, the visitor E is not imaged by the collective entrance camera 101. Therefore, it is determined that the feature (for example, feature E) of the person (visitor E) extracted by the handset camera 302 does not match the feature A of the visitor A captured by the collective entrance camera 101. Accordingly, the determination unit 804 determines that the person (visitor E) is the unknown person Y. The feature of the unknown person Y is recorded as visitor information in the data table in association with, for example, “unknown person Y: feature E” (see step S121, FIG. 9A).

  In addition, unlike the flow line of the visitor A, the flow line of the unknown person Y is determined not to match the predicted route R1 (the predicted route of the visitor A captured by the collective entrance camera 101) in step S123. . Thereby, the determination unit 804 determines that the unknown person Y is an intruder (step S125).

  According to the above operation, by comparing the flow line of an unknown person (visitor D, E) that was not captured by the collective entrance camera 101 with the predicted route R1 of the visitor A captured by the collective entrance camera 101, It is determined whether the unknown person is an intruder. For this reason, it is determined that an unknown person (visitor D) who acts together with the visitor A imaged by the collective entrance camera 101 (the flow line coincides) is a companion of the visitor A, etc. Can be determined not to be an intruder. Further, it is possible to determine that an unknown person (visitor E) who takes a completely different action from visitor A (the flow line does not match) is an intruder who enters the apartment together with visitor A. Thereby, the security with respect to an intruder can further be improved.

  In addition, this invention is not limited to embodiment mentioned above, A deformation | transformation, improvement, etc. are possible suitably. In addition, the material, shape, dimension, numerical value, form, number, location, and the like of each component in the above-described embodiment are arbitrary and are not limited as long as the present invention can be achieved.

  For example, at least one of the extraction unit 801, the predicted route determination unit 802, the flow line estimation unit 803, the determination unit 804, and the recording unit 805 of the controller 80 is a server device 81 installed in an apartment or a server device on an external cloud It is good also as a structure arrange | positioned in either. With this configuration, the processing load on the controller 80 can be distributed, and stable operation of the entire system including the controller 80 can be maintained.

1: Intercom system for collective housing, 10: collective entrance machine, 20: living room master machine, 30: entrance slave machine, 40: collective entrance door, 50: entrance door, 60: control room master machine, 70: surveillance camera, 80 : Controller, 101: Collective entrance camera, 102: Input unit, 301: Human sensor, 302: Slave camera, 801: Extraction unit, 802: Predicted route determination unit, 803: Flow line estimation unit, 804: Judgment unit 805: Recording unit, 806: Notification unit

Claims (11)

A collective entrance machine having a collective entrance camera and an input unit for inputting room number information;
A human presence sensor that senses a person, a slave camera that is activated based on the output of the human sensor, and an entrance slave that is installed at the entrance of each room;
When the room number information is input to the input unit, the entrance imaging data captured by the collective entrance camera and the room number information are transmitted from the collective entrance machine, and the human sensor detects a person A controller that transmits the slave image data captured by the slave camera to the entrance slave unit;
With
An extraction unit for extracting characteristics of a person included in the entrance imaging data;
A predicted route determination unit that determines a predicted route of the person based on the room number information;
A recording unit for recording the feature and the predicted route in association with the person;
When the child image data transmitted from the entrance child device of each room includes the characteristics of a person imaged by the collective entrance camera, the image was captured by the child camera based on the child device image data. A flow line estimation unit for estimating a flow line of a person,
A determination unit that determines whether the person imaged by the slave camera is an intruder by comparing the predicted path and the flow line;
Having
Intercom system for collective housing. When the entrance door of the living room is opened and closed, opening / closing information indicating that the entrance door has been opened and closed is transmitted from the living room to the controller,
The extraction unit extracts a feature as a person who is out of the person imaged by the child camera in the living room that transmitted the opening / closing information,
The recording unit records the opening / closing information and the characteristics of the outing person in association with the outing person,
2. The intercom system for an apartment house according to claim 1, wherein the determination unit determines that the person imaged by the slave camera is not an intruder when the person imaged by the slave camera is the outing person. .   The said determination part determines that the person imaged with the said subunit | mobile_unit camera is an intruder when the characteristics of the person imaged with the said entrance camera are not contained in the said subunit | mobile_unit imaging data. Or the intercom system for apartment houses of Claim 2. When the characteristics of the person imaged by the collective entrance camera is not included in the slave device imaging data,
The extraction unit extracts a feature as an unknown person included in the handset imaging data,
The predicted route determination unit determines the predicted route of the unknown person based on the predicted route of the person imaged by the collective entrance camera,
The flow line estimation unit estimates the flow line of the unknown person based on the handset imaging data,
The apartment unit according to claim 1, wherein the determination unit determines whether the unknown person is an intruder by comparing the unknown person's predicted route and the flow line of the unknown person. Intercom system for.   5. The set according to claim 1, wherein the recording unit records at least one of the handset imaging data including the person determined to be an intruder and the entrance imaging data. 6. Residential intercom system.   The said control machine is provided with the alerting | reporting part which alert | reports a warning sound, when the said judgment part judges the person imaged with the said subunit | mobile_unit camera as an intruder. The intercom system for apartment buildings described.   The said control machine is provided with the notification part which notifies the result of the said judgment to the management room main | base station of a housing complex when the said judgment part judges the person imaged with the said subunit | mobile_unit camera as an intruder. The intercom system for collective housing as described in any one of Claim 6.   When the determination unit determines that the person imaged by the slave camera is an intruder, the controller transmits the slave unit image data including the person imaged by the slave camera to the management room master unit The intercom system for collective housing according to claim 7, wherein the intercom system is transmitted to. It has a surveillance camera installed in the apartment house,
The intercom system for an apartment house according to any one of claims 1 to 8, wherein the determination unit determines whether or not the person imaged by the monitoring camera is an intruder.   At least one of the extraction unit, the predicted route determination unit, the recording unit, the flow line estimation unit, and the determination unit includes the controller, a server device installed in an apartment house, and a server device on an external cloud. The intercom system for an apartment house according to any one of claims 1 to 9, which is disposed at any one of the locations. The determination unit, when the flow line of the person imaged by the slave camera deviates from the predicted path, when the person imaged by the slave camera returns to the predicted path within a predetermined time, The intercom system for collective housing according to any one of claims 1 to 10, wherein the person is determined not to be an intruder.

Images