JP7005648B2 - Elevator usage log output system and elevator usage log output method - Google Patents

Description

The present invention relates to an elevator use log output system for accurately grasping the movement information of a user who uses an elevator, and an elevator use log output method.

In relatively large buildings, multiple elevators are installed in order to improve the transportation capacity of users by elevators, and a system is introduced to select and service the most suitable car when registering a call at the landing. .. Furthermore, as the scale of the building increases, the number of elevators installed side by side also increases, and these multiple elevators are appropriately controlled by the group management control device to improve services such as reducing waiting time for users. ing.

In an elevator using such a group management control device, in order to appropriately control each elevator, the behavior of the user and the like are measured to control the elevator. For example, in Japanese Patent Application Laid-Open No. 2010-254391 (Patent Document 1), Japanese Patent Application Laid-Open No. 2003-221174 (Patent Document 2), Japanese Patent Application Laid-Open No. 2006-21852 (Patent Document 3), the following system is used. Proposed.

In the elevator system shown in Patent Document 1, a detector for detecting a retracted event is provided around the landing door, and a history recording unit for recording the retracted event by this detector is provided. A detection number aggregation unit that counts the number of event detections by the history recording unit is provided, and the opening and closing of the landing door is controlled based on the number of event detections in a predetermined period. According to this, it is possible to prevent the event of being pinched by the landing door and the event of being drawn into the door pocket.

Further, in the elevator system shown in Patent Document 2, a surveillance camera is provided in the car, and a camera image is transmitted to an elevator maintenance company at a predetermined time to confirm the user. According to this, the last person leaving the building and the person entering the building at night can be easily confirmed from a remote place, and the security function of the building can be improved.

Further, in the elevator system shown in Patent Document 3, a video recording means for recording the video in the car taken by the surveillance camera is provided, and when a signal relating to an event having a high possibility of crime is received, the reception time is reached. Along with the data and the identification data representing each event, the video in the car in the predetermined time zone recorded in the video recording means is saved. According to this, the recording capacity of the image in the car can be reduced, and the image at the time of the event can be searched quickly.

Japanese Unexamined Patent Publication No. 2010-254391 Japanese Patent Application Laid-Open No. 2003-221174 Japanese Unexamined Patent Publication No. 2006-21852

By the way, in an elevator equipped with this kind of group management control device, in order to operate the elevator efficiently, it is often required to perform simulated operation by simulation in advance. For this purpose, it is necessary to accurately grasp the movement information of the user (movement of the building upward or downward, the number of users, the boarding machine, the waiting time, the departure floor, the arrival floor, etc.). There is. If the movement information of the user can be accurately grasped in this way, the accurate parameter data required for the preliminary simulation for performing the group management control can be obtained, so that the accuracy of the simulation can be improved. Therefore, in the group management control device, it is strongly required to accurately grasp the movement information of the user.

An object of the present invention is to provide an elevator use log output system capable of accurately grasping user's movement information, and an elevator use log output method.

The feature of the present invention is to set and assign discrimination information and ID information for discriminating a user to each user from an image of a landing camera that shoots the landing of an elevator or a camera in the car that shoots the inside of a car. Based on this discrimination information and ID information, the movement of the user is traced, and at least the boarding floor and the disembarking floor of the user are associated and stored, and the movement information of the user is grasped.

According to the present invention, since the movement information of the user can be accurately grasped, for example, accurate parameter data required for prior simulation for performing group management control can be obtained, so that the accuracy of the simulation can be improved.

It is a block diagram which shows the outline of the elevator operation management system and the surveillance camera control system which become 1st Embodiment of this invention. It is an external view which looked at the landing from diagonally above. It is explanatory drawing which shows the user detection range set for each boarding machine. It is explanatory drawing explaining the movement of the user of the boarding machine from the 4th floor to the 1st floor. It explains the movement of the user, and is an explanatory diagram showing the landing on the 5th floor before the user gets on the boarding machine. It explains the movement of the user, and is an explanatory diagram showing the landing on the 5th floor after the user gets on the boarding machine. It explains the movement of a user, and is an explanatory diagram showing a landing on the 7th floor where one user got off. It explains the movement of users, and is an explanatory diagram showing a landing on the 8th floor where two users got off. It is an explanatory diagram showing the platform on the 9th floor where one user got off, which explains the movement of the user. It is explanatory drawing which shows the landing on the 9th floor before one new user boarded. It is a flowchart explaining the method of acquiring the movement information of the user of the elevator which becomes embodiment of this invention. It is a detailed control flow of the control step S12 shown in FIG. It is a detailed control flow of the control step S13 shown in FIG. It is a detailed control flow of the control step S20 shown in FIG. It is a figure which shows the specific data of the log output executed in the control step S19 shown in FIG. It is a figure which shows the other concrete data of the log output executed in the control step S19 shown in FIG.

Next, an embodiment of the present invention will be described in detail with reference to the drawings, but the present invention is not limited to the following embodiments, and various modifications and applications are included in the technical concept of the present invention. Is also included in the range.

FIG. 1 is a configuration diagram showing an outline of an elevator controlled by group management according to the first embodiment of the present invention.

The elevator operation management system 10 that functions as a group management control device is connected to the boarding machine elevator control systems 11A to 11N that control each boarding machine of a plurality of elevators, and the boarding machine elevator control systems 11A to 11N operate the elevator. It has a function of receiving a control command from the management system 10 and actually operating the boarding machine. For example, it controls the electric motor of the hoisting machine of the boarding machine, the brake mechanism of the hoisting machine, the door opening / closing electric motor, and the like.

Further, the elevator operation management system 10 is connected to the landing elevator service request device 13, the building management system 14, the public institution management system 15, and the surveillance camera control system 16 via the communication network 12. Since the landing elevator service request device 13, the building management system 14, and the public institution management system 15 are not related to the present embodiment, detailed description thereof will be omitted. Further, the system configuration shown in FIG. 1 is operated as a private system constructed for each building or for each of a plurality of buildings.

The elevator operation management system 10 according to the present embodiment is composed of a learning unit 10A, a receiving unit 10B, a floor-based number evaluation unit 10C, a comprehensive evaluation unit 10D, and an allocation command unit 10E, each of which is a computer. It can be realized as a control function.

The receiving unit 10B is connected to the communication network 12, and various related information from the surveillance camera control system 16 is input. The received various related information is sent to the learning unit 10A, and the various related information is learned by executing the rewriting process or the like.

Further, the received various related information is sent to the floor-specific number evaluation unit 10C, where a predetermined evaluation calculation process is executed. The evaluation calculation result is sent to the comprehensive evaluation unit 10D, and the comprehensive evaluation calculation is executed together with other evaluation calculation parameters. The operation allocation information of the boarding machine calculated by the comprehensive evaluation unit 10D is sent to the allocation command unit 10E, and the allocation command unit 10E sends a control command to the corresponding boarding machine elevator systems 11A to 11N, and the boarding unit elevator system. Have 11A-11N perform a predetermined function.

On the other hand, the surveillance camera control system 16 which is a feature of the present embodiment has an image input processing unit 16A, a detection range setting processing unit 16B, a user detection processing unit 16C, an ID setting processing unit 16D, and a floor-specific user of the surveillance camera. It is composed of a detection processing unit 16E, a movement information output processing unit 16F, and a log output processing unit 16G, which can also be realized as a control function of a computer.

The detection range setting processing unit 16B has a function of setting the user detection range of the user detected by the following user detection processing unit 16C, and the user detection range can be set in any range. For example, in front of the landing door in front of the boarding machine of the landing, a semicircular detection range having a predetermined radius can be set, or a rectangular detection range can be set.

The user detection processing unit 16C obtains an image feature amount for detecting a person (= user) from an image taken by a surveillance camera input to the image input processing unit 16A, and determines a user from this feature amount. It has a function to identify and extract. Alternatively, it has a function of identifying and extracting a user by comparing a model of a person's head image or an overall image with a captured image.

In any case, the user detection processing unit 16C has a function of identifying and extracting a user from an image taken by a surveillance camera. Further, the user detection processing unit 16C has a function of storing detected individual user discrimination information, for example, an image feature amount of an individual user. This makes it possible to identify and identify users who have moved to other floors. In the following, the image feature amount for discriminating the user will be described as the discrimination information, but if the information identifies the user other than the image feature amount, it will be treated as the discrimination information.

The ID setting processing unit 16D has a function of setting and assigning ID information to all users detected by the user detection processing unit 16C. For example, if the ID information is associated with the face or body discrimination information of the user, the ID information can be traced by the image analysis of the user. It is also possible to link the identification information of individual users with the ID information by other methods. In addition, it is desirable to encrypt the identification information, ID information, etc. of individual users in order to ensure confidentiality.

Furthermore, it is possible to set conditions for releasing the associated user and ID. These may be values that are fixed when the system is delivered, or the release conditions may change depending on the communication request of various linked systems. This is because it is necessary to store the characteristic points of the user in the computer in order to associate the user with the ID, and therefore it is necessary to consider confidentiality more in the usage scene of the system.

The user detection processing unit 16E for each floor detects the movement information of each user based on the user's discrimination information and the ID information for each floor. The user's movement information detected here is, for example, based on the user's discrimination information and ID information, the time when the ID information is generated and its floor, the boarding machine number, the user's boarding time and its floor, and the use. The time of getting off the person and the floor of the person.

Such movement information can be acquired by setting and assigning discrimination information and ID information by image analysis of the user, and can be traced by a surveillance camera on each floor or a surveillance camera in the car. As a result, the above-mentioned movement information of the user can be accurately obtained.

The movement information output processing unit 16F has a function of storing the movement information of the user for each floor obtained by the user detection processing unit 16E for each floor in a separately rewritable storage area so that the log can be output. I have.

The log output processing unit 16G has a function of outputting a log of the user's movement information stored in the movement information output processing unit 16F. Therefore, the movement information of the user output to the log can be effectively used as the parameter data of the preliminary simulation for performing the group management control.

Next, the details of the surveillance camera control system 16 which is a feature of the present embodiment will be described with reference to FIGS. 2 to 11. 2 to 5 are diagrams for explaining the concept of the present embodiment, and FIGS. 6 to 11 are diagrams for explaining the specific embodiment thereof.

FIG. 2 shows a landing on a certain floor (for example, the 4th floor), and a landing monitoring camera 17 is installed at an arbitrary position on the landing. As the landing surveillance camera 17, a wide-angle camera capable of photographing the entire landing is used. In addition, four elevators are installed at this platform, and the boarding units 4A to 4D are operated respectively.

Needless to say, the landing surveillance cameras 17 are also arranged on each floor other than this floor. The landing surveillance camera 17 on each floor is configured as a network camera, and is comprehensively managed and controlled by the surveillance camera control system 16, whereby the user can be traced. Then, a user is waiting at the landing in front of the boarding machine 4B, which is operated upward, and the boarding machine 4D, which is operated downward. Four downward user Pdns are waiting at the boarding machine 4B, and six upward user Pups are waiting at the boarding machine 4D.

FIG. 3 shows a user detection range for performing image analysis for detecting a user for each boarding machine. In front of the landing doors 18A to 18D of each boarding unit 4A to 4D, the user detection range 19A to 19D is set in a semicircular shape with a predetermined radius centered on the vicinity of the center of each of the landing doors 18A to 18D. Therefore, the user detection range 19A to 19D is set for each of the boarding units 4A to 4D.

At present, since the user is waiting in the boarding machine 4B and the boarding machine 4D, it is possible to detect the user existing in the user detection range 19B and the user detection range 19D by image analysis. The user detection range can be arbitrarily set by the image input processing unit 16A and the detection range setting processing unit 16B, and can be set to a rectangular shape or any other shape in addition to the semicircular shape. Is also possible.

For all users detected in the user detection range, image feature points, which are user discrimination information, are extracted by image analysis, and ID numbers are set so that they are associated with the image feature points of individual users. , Granted. Therefore, if there is a user whose image feature points of the user taken by another landing surveillance camera 17 match, the movement trajectory of the user can be estimated from the ID number.

Here, for the user Pdn-1 located in the overlapping region where the adjacent user detection ranges 19A to 19D overlap, a state occurs in which it cannot be determined which user detection range belongs to. In this case, by image analysis, feature points such as the user's face and shoulders can be extracted, and the user detection range and the boarding machine thereof can be determined from the posture and direction in which the user is facing. Alternatively, since the movement locus of the user can be estimated from the temporal change of the image, the user detection range and the boarding machine can be determined by determining which boarding machine the user has moved to.

In addition, even when waiting for the elevator, the position coordinates of the elevator are set in advance, or the elevator door is detected by the image taken from the landing, and the position coordinates are automatically set. As described above, it is possible to determine from which elevator position coordinates the detected user's posture or direction is heading, and to determine which boarding machine the elevator is waiting for.

Therefore, in FIG. 3, the user Pdn-1 is detected in the user detection range 19D as a user who gets on the boarding unit 5D. The user detection range can be determined by the same processing in the overlapping area of other user detection ranges. Needless to say, the user existing in the overlapping area is also set and assigned an ID number associated with the image feature point which is the discrimination information.

In such a state, it is possible to acquire movement information such as an image feature point, an ID number, a time when the ID number is assigned, a boarding floor, a boarding machine, etc., which are discrimination information of individual users. Further, when the user disappears from the user detection range, the boarding time information can be acquired as the time when the user boarded the corresponding boarding machine.

FIG. 4 shows a moving state of a user who gets on the boarding machine 4B shown in FIG. 3 and moves downward. Landing surveillance cameras 17-4 to 17-1 are installed on each floor, and image information of each landing monitoring camera 17 is sent to the surveillance camera control system 16 for comprehensive management and control. The "n" in the reference number 17-n of the surveillance camera means the floor number. Similarly, in the boarding machine 4B, the in-car surveillance camera 20 is installed in the car 21, and the image information of the surveillance camera 20 is also sent to the surveillance camera control system 16 for comprehensive management and control.

Next, the moving state of the user will be described with reference to FIGS. 5A to 5E. FIGS. 5A to 5E show a case where the car 21 moves upward as opposed to FIG.

First, the user waiting on the 5th floor is waiting for the arrival of the boarding machine 4B as shown in FIG. 5A. In this case, there are four users, users Pa to Pd. Therefore, at this time, by analyzing the image from the landing monitoring camera 17-5, the discrimination information of the individual users of the users Pa to Pd within the user detection range 19B is extracted. Further, the ID number is set and assigned so as to be associated with the discrimination information of the users Pa to Pd, and at the same time, the set time of the ID number is measured and stored. Then, when the boarding machine 4B arrives, the users Pa to Pd board the boarding machine 4B as shown in FIG. 5B. At this time, when the users Pa to Pd disappear from the user detection range 19B, it is determined that the users Pa to Pd have boarded the boarding unit 4B, and the boarding floor and the boarding time are measured and stored.

When the users Pa to Pd get on the boarding machine 4B, the image of the users Pa to Pd is sent to the surveillance camera control system 16 by the surveillance camera 20 in the car of the car 4B, and the images of the users Pa to Pd are transmitted from this image to the surveillance camera control system 16. Discrimination information is requested and compared with the user's discrimination information sent from the landing surveillance cameras 17-5. If the comparison results are the same, it is recognized that all the users Pa to Pd at the landing have boarded the boarding machine No. 4B. On the other hand, if the comparison results are different, it is recognized that some of the users who do not match have moved using the stairs or the like.

Next, when the boarding machine 4B rises and stops on the 7th floor, the user Pd gets off as shown in FIG. 5C, but the image of the user Pd is transferred to the surveillance camera control system 16 by the landing monitoring camera 17-7. It is sent, and the discrimination information of the user Pd is obtained from this image, and is compared with the discrimination information of the user Pd of the image sent from the landing surveillance cameras 17-5. If the comparison results are the same, it is recognized that the user Pd has disembarked on the 7th floor, and the disembarkation floor, the disembarkation time, and the like are stored.

Since the user Pd disappears from the image of the in-car surveillance camera 20 of the car 4B, it can be recognized that the user Pd got off on the 7th floor. Therefore, the accuracy of the user's recognition can be improved by matching the discrimination information by the landing monitoring camera 17-7 with the discrimination information by the in-car surveillance camera 20 of the car. Of course, it goes without saying that the accuracy of recognition can be improved by executing the same processing on the following floors.

Next, when the boarding machine 4B rises and stops on the 8th floor, the users Pa and Pb get off as shown in FIG. 5D. It is sent to the control system 16, and the discrimination information of the users Pa and Pb is obtained from this image, and is compared with the discrimination information of the users Pa and Pb of the image sent from the landing surveillance cameras 17-5. If the comparison results are the same, it is assumed that the users Pa and Pb have disembarked on the 8th floor, and the disembarkation floor, the disembarkation time, and the like are stored. Since the users Pa and Pb disappear from the image of the in-car surveillance camera 20 of the car 4B, it can be recognized that the users Pa and Pb have disembarked on the 8th floor.

Next, when the boarding machine 4B rises and stops on the 9th floor, the user Pc gets off as shown in FIG. 5E, but the image of the user Pc is transferred to the surveillance camera control system 16 by the landing monitoring camera 17-9. It is sent, and the discrimination information of the user Pc is obtained from this image, and is compared with the discrimination information of the user Pc of the image sent from the landing surveillance cameras 17-5. If the comparison results are the same, it is assumed that the user Pc has disembarked on the 9th floor, and the disembarkation floor, the disembarkation time, and the like are stored. Since the user Pc disappears from the image of the in-car surveillance camera 20 of the car 4B, it can be recognized that the user Pc got off on the 9th floor.

Next, when a new downward user Pe appears at the landing on the 9th floor, the user Pe exists in the user detection range 19C as shown in FIG. 5F, and the identification information of the user Pe is extracted. And ID information is set and assigned, and at the same time, the set time of the ID number is measured and stored. After that, the movement of the user Pe is traced and the movement information is acquired.

Based on this idea, a specific embodiment will be described below. FIG. 6 is a flowchart showing a control flow by the computer of the above-mentioned processing executed by the surveillance camera control system 10. This control flow is activated at the activation timing at predetermined time intervals.

<< Step S10 >>
In step S10, the user detection range setting process is executed, and the user detection ranges 19A to 19D as shown in FIG. 3 are set based on the input image of the landing. Then, from the images in the user detection ranges 19A to 19D, the extraction of the user for each boarding machine is executed by the following step S11. When the setting process of the user detection range is completed, the process proceeds to step S11.

≪Step S11≫
In step S11, the user detection process is executed, the images in the user detection ranges 19A to 19D set in step S10 are analyzed, and the extraction of individual users for each boarding machine is executed. It is something to do. In the extraction of individual users, an image feature amount for extracting a user is obtained from an image taken by a landing surveillance camera 17, and an individual user is specified and extracted from this feature amount. Since this feature amount is used as discrimination information for individual users, it is stored in a storage area (not shown).

Here, when the images in the user detection range 19A to 19D are analyzed, if the user exists in the overlapping area where the user detection range overlaps, it is set to which user of the user detection range. It may not be possible to determine whether to do so. Therefore, when the user detection process is completed, the process proceeds to step S12 to execute the setting process of the user existing in the overlapping area.

≪Step S12≫
In step S12, the user detection range of the user existing in the overlapping area is reset, and a process called human vector detection is executed here. This process determines which user detection range the user Pdn-1 (see FIG. 3) existing in the overlapping area belongs to and resets it. By image analysis, the user's face, shoulders, etc. It is possible to extract the feature points of the above and determine the user detection range and the boarding machine from the direction and posture in which the user is facing. In addition to this, since the movement locus of the user Pdn-1 can be estimated from the temporal change of the image, the user detection range and the boarding machine are determined by determining which boarding machine the user has moved to. can do. In this way, it is possible to set the user detection range of the user existing in the overlapping region from at least one or more information of the posture and direction of the user or the movement locus. The human vector detection process will be described in detail with reference to FIG. When the person vector detection process is completed, the process proceeds to step S13.

≪Step S13≫
In step S13, ID numbers are set and assigned to the individual discrimination information of all the users extracted in steps S11 and S12. By associating the ID information with the discrimination information consisting of the features of the user's face and body, it is possible to trace the ID information of the moving user by the image analysis of the user. In this step S13, the ID number associated with the identification information of each user is stored.

For example, an ID number of "0001" is set for the user Pa as shown in FIG. 4, an ID number of "0002" is set for the user Pb, and an ID number of "0003" is set for the user Pc. The ID number of "0004" can be set in the user Pd. Therefore, for example, when the user Pa moves from a certain departure floor to a different arrival floor, since the discrimination information of the user Pa is acquired on the departure floor, the image of the user Pa on the arrival floor. If the same discrimination information is obtained in the analysis, the user Pa can be traced by the ID number "0001". This ID assignment process will be described in detail with reference to FIG.

It is also possible to link the identification information of individual users with the ID information by other methods. In addition, it is desirable to encrypt the identification information, ID information, etc. of individual users in order to ensure confidentiality. When the process of assigning the ID number is completed, the process proceeds to step S14.

<< Step S14 >>
In step S14, an individual user reaches the user detection range of the landing, sets an ID number, and stores the assigned time (occurrence time). Therefore, the occurrence time differs depending on the ID number, and for example, the departure time distribution of the user for each floor can be acquired.

≪Step S15≫
In step S15, it is detected whether or not the user has boarded the arriving boarding machine. As this detection method, it can be detected based on the disappearance of the user from the image. That is, when the user disappears from the user detection range, it is determined that the user has boarded the boarding machine. Of course, in this case as well, since the user identification information and the ID information are acquired, it is possible to obtain the boarding information of each individual user.

Further, it is possible to obtain the boarding information of the user from the image of the in-car monitoring camera 20 provided in the car 21 instead of judging from the image of the landing monitoring camera 17. As described above, since the user identification information and the ID number are acquired by the landing surveillance camera 17, the use that has been waiting at the landing can be compared with the user identification information by the image of the in-car surveillance camera 20. It is possible to determine whether or not a person has boarded. When the detection process of the passenger is completed, the process proceeds to step S16.

≪Step S16≫
In step S16, the time when the user boarded the boarding machine (boarding time) is stored. In this case as well, since the user identification information and the ID number are acquired, as shown in step S15, the boarding time for each user can be acquired from the boarding information detected for each user. When the boarding time is acquired, the process proceeds to step S17.

≪Step S17≫
In step S17, when the boarding machine on which the user has boarded reaches the target disembarkation floor, the disembarkation information of the user is detected. The user's disembarkation detection can be detected by the landing surveillance camera 17 on the arrival floor. That is, the user gets off at the arrival floor, but the image of the user is sent to the surveillance camera control system 16 by the landing surveillance camera 17 on the arrival floor. In the surveillance camera control system 16, discrimination information of the user who got off is obtained from the transmitted image, and is compared with the discrimination information of the user of the image sent from the landing surveillance camera 17 on the departure floor. If the comparison results are the same, it is recognized that the user who got on from the departure floor got off, and the ID number, the getting off floor, etc. are stored. Since the user disappears from the image of the surveillance camera 20 in the car, it can be estimated that the user has disembarked. When the detection process of the disembarking user is completed, the process proceeds to step S18.

≪Step S18≫
In step S18, the time when the user gets off the boarding machine (getting off time) is stored. In this case as well, since the user identification information and the ID number are acquired, as shown in step S17, the time of disembarkation for each user can be acquired from the disembarkation information detected for each user. When the disembarkation time is acquired, the process proceeds to step S19.

≪Step S19≫
In step S19, a log storage area in which the user's movement information can be rewritten so that the movement information (movement history) of individual users acquired by executing each of the above-mentioned control steps can be output as a log. I remember it. In this case, the "user identification information", the "ID number" associated with the "user identification information", the "occurrence time (set time)" of the ID number, and the "occurrence floor number" in which the ID number was generated, The "boarding time" and "boarding floor number" of getting on the boarding machine and the "getting off time" and "getting off floor number" of getting off from the boarding machine are stored. In addition to this, information on the user detection range and the boarding machine is also stored, so that it can be stored in the log storage area as needed. When the user's movement information is stored in the rewritable log storage area, the process proceeds to step S20.

≪Step S20≫
In step S20, the ID number of the individual user and the discrimination information associated with the ID number are released (erased), and only the substantial movement information of the user is left as a history. When the step S20 is completed, the control flow of FIG. 6 is terminated, and the patient waits for the next activation timing. In this way, since the movement information of the user can be accurately grasped, the accuracy of the preliminary simulation for performing the group management control can be improved. This ID release process will be described in detail with reference to FIG.

Next, a specific example of step S12 shown in FIG. 6 will be described with reference to FIG. 7. When step S11 shown in FIG. 6 is completed, steps S30 to S34 shown in FIG. 7 are executed, and the process proceeds to step S13 again.

<< Step S30 >>
In step S30, it is determined from the image of the camera whether or not a new user is detected at the landing. Since it is possible to identify individual users by image analysis, when a new user appears at the landing for users before the present time, it is possible to identify the new user by extracting this. In step S30, if a new user is not extracted, the process exits to the end, and if a new user is extracted, the process proceeds to step S31.

≪Step S31≫
In step S31, as shown in FIG. 3, it is determined whether or not the new user Pdn-1 exists in the overlapping region of the user detection range 19D and the user detection range 19C. Then, when it is determined that the new user Pdn-1 exists in the overlapping area, the process proceeds to step S32, and when it is determined that the new user Pdn-1 does not exist in the overlapping area, step S33. Move to.

<< Step S32 >>
In step S32, the person vector processing (step S12) shown in FIG. 6 determines the user detection range to which the new user belongs, and thereby the boarding machine is also determined. This is quantified as the detected position coordinates of the user and an angle with respect to the direction in which the user is facing with respect to an arbitrary direction. For example, taking the user Pdn-1 in FIG. 3 as an example, in a layout in which the landing plane is viewed from above, the angle is set to 0 ° with respect to the angle perpendicular to the upper side in the figure. From there, the position coordinates of the elevator and the position coordinates of the user are used to detect which elevator is facing.

For example, simply, if it is 1 ° to 90 °, it is waiting for Unit 1, if it is 91 ° to 180 °, it is waiting for Unit 2, if it is 181 ° to 270 °, it is waiting for Unit 3, and if it is 271 ° to 0 °, it is 4 It may be a method of waiting for a unit or a method in which a position area of an elevator is set and the angle for detecting each of the above-mentioned waiting units is changed according to the position coordinates of the user. In the case of the example of FIG. 3, if the orientation of the user is detected and the user of the user Pdn-1 is set to 120 °, it is determined that the user is waiting for Unit 2.

In this way, resetting the user detection range sets the waiting direction of each boarding machine, detects the waiting direction of the user in the overlapping area from the image of the surveillance camera, and compares it with the waiting direction of the boarding machine. , The user detection range of the users existing in the overlapping area can be set.

≪Step S33≫
Since it is determined in step S31 that the new user clearly belongs to the user detection range, in step S33, the boarding machine is determined based on this user detection range.

<< Step S34 >>
In step S34, the boarding machine number is assigned to the user in the overlapping area and the user extracted in the user detection range. On the other hand, even when a new user appears and does not exist in the overlapping area but exists in the user detection range, the boarding machine number is assigned to the user extracted in the user detection range. When this process is completed, the process proceeds to step S13 of FIG. 6 to continue the operation of the control flow shown in FIG.

Next, a specific example of step S13 shown in FIG. 6 will be described with reference to FIG. When step S12 shown in FIG. 6 is completed, steps S40 to S41 shown in FIG. 8 are executed, and the process proceeds to step S14 again.

<< Step S40 >>
In step S40, it is determined whether or not the discrimination information of the user existing in the user detection range is detected for the first time. In this case, since the user discrimination information existing in the user detection range before the present time is already stored in the storage area of the surveillance camera control system 16, the stored discrimination information and the discrimination detected this time are performed. By comparing the information, it can be determined whether or not the user is detected for the first time.

If it is not the discrimination information detected for the first time, the end is exited, and if the discrimination information is detected for the first time, the process proceeds to step S41.

≪Step S41≫
In step S41, a new ID number is assigned to the user's discrimination information detected for the first time. Then, since the user's discrimination information and the ID number are associated and stored, it is possible to trace the movement state of the user to which the ID number is assigned. When this process is completed, the process proceeds to step S14 of FIG. 6 to continue the operation of the control flow shown in FIG.

Next, a specific example of step S20 shown in FIG. 6 will be described with reference to FIG. When step S19 shown in FIG. 6 is completed, steps S50 to S51 shown in FIG. 9 are executed to move to the end.

<< Step S50 >>
In step S50, it is determined whether or not the ID number release process is necessary. In the present embodiment, a case where the user's movement information (movement history) is traced over a predetermined time and a case where the user's movement information (movement history) is not traced are set, and as an example, it is set in "daily units". Then, if the release letter condition is set to "daily unit", the process proceeds to step S51, and if it is not set, the process proceeds to step S52.

≪Step S51≫
In step S51, the discrimination information (feature amount) and the ID number are not released but are linked and held until the date and time reaches 0:00. This makes it possible to trace the movement state of the user over the time of the day.

<< Step S52 >>
In step S52, when the predetermined release condition is satisfied within the time less than "daily unit", the ID number assigned to the individual user and the discrimination information associated with the ID number are released (erased). Will be done. As mentioned earlier, what is released is the ID number given to each user and the discrimination information associated with it, and the "occurrence time (set time)" of the other ID numbers. , "Occurrence floor number" where the ID number was generated, "Boarding time" and "Boarding floor number" when boarding the boarding machine, "Getting off time" and "Getting off floor number" when getting off from the boarding machine, etc. are simulated. It is retained as parameter data of.

Here, the release conditions set in step S52 include (1) the release condition is that the user gets off the boarding machine, and (2) the time such as one lap time in the traffic calculation used for the elevator installation plan. The release condition is the expiration of the used time. (3) If there is a first elevator group and a second elevator group, the first elevator group to the second elevator group, or the second elevator group, assuming a transfer. The release condition is that you finally get off when you transfer from the elevator group to the first elevator group (for example, if you transfer from a low-rise to a high-rise, the release condition is when you get off the high-rise elevator). Etc., so it is sufficient to set appropriate release conditions.

FIG. 10 shows a specific example of the log aggregation process executed in step S19, and is a log output when the ID release condition is the case of getting off from the boarding machine.

The log output includes "user identification information" representing the user, the "ID number" associated with the user, the "occurrence time" of the ID number, and the "occurrence floor number" in which the ID number was generated. , The "boarding time" and "boarding floor number" of getting on the boarding machine, and the "getting off time" and "getting off floor number" of getting off from the boarding machine are stored.

For example, for the user Pa, the ID number is set to "0001", this ID number is set to "8:00:01" on the "5th floor", and the user Pa is set to "5th floor". At the landing, the boarding machine boarded at "8:01:05", and the user Pa can acquire information that the boarding machine got off at "8:01:20" at the boarding point on the "8th floor". The same applies to other users, and it is possible to improve the accuracy of the preliminary simulation for performing group management control from the movement information of these users.

When the ID release process is executed, the user identification information and the "ID number" associated with the user identification information are deleted and become general user information (for example, A, B, C ...). Will be replaced. Since the simulation does not require the personal information of the user, there is no problem even if the personal information such as the discrimination information of the user is deleted, and it is also desirable from the viewpoint of protecting the personal information.

FIG. 11 shows a specific example of the log aggregation process executed in step S19, and is a log output when "daily unit" is set as the ID release condition. The log output is basically the same as in FIG. 10, but since the ID release condition is "daily unit", the movement information is acquired in "daily unit" as shown in the user Pc.

For example, in the first detection process, the ID number of the user Pc is set to "0003", the ID number is set to "8:00:04" on the "fifth floor", and the user Pc is set to "8:00:04". At the "5th floor" platform, board the boarding machine at "8:01:06", and the user PC will get off at the "9th floor" platform at "8:01:55" for office work. You can get the information that you went to the place.

In addition, even after the user Pc has worked in the office nine times and the time has passed, the ID number of the user Pc is retained as "0003", and this ID number is "10:" on the "9th floor". It was rediscovered at "10:25", and the user Pc boarded the boarding machine at "10:10:36" at the boarding point on the "9th floor", and the user Pc boarded the boarding point on the "1st floor". You can get the information that you got off at "10: 11: 03" from the boarding machine. The same applies to other users, and it is possible to improve the accuracy of the preliminary simulation for performing group management control from the movement information of these users. Furthermore, by setting the ID release condition, it is possible to create data for maintaining the accuracy of the simulation in any usage scene in consideration of privacy.

In this embodiment, the above-mentioned "user discrimination information" representing the user, the "ID number" associated with the "ID number", the "occurrence time" of the ID number, and the "occurrence floor" in which the ID number is generated are generated. The "floor number", the "boarding time" and "boarding floor number" for getting on the boarding machine, and the "getting off time" and "getting off floor number" for getting off from the boarding machine are stored, but if necessary. It is also possible to set other storage items.

The above-described embodiment shows an example in which the images of the landing surveillance camera 17 provided in the landing on each floor and the in-car surveillance camera 20 provided in the car are comprehensively managed and controlled by the surveillance camera control system 16. However, the movement information of the user may be acquired using only the landing monitoring camera 17 provided at the landing on each floor, or the user may use only the car monitoring camera 20 provided in the car. It is also possible to acquire the movement information of.

When the movement information of the user is acquired by using only the landing monitoring camera 17 provided at the landing on each floor, it can be basically realized by the same control as the control flow shown in FIG. In this case, the boarding user detection process in step S15 and the disembarking user detection process in step S17 detect the behavior of the user from the image of the boarding area monitoring camera 17 and the image of the boarding area monitoring camera 17 getting off. It can be carried out at. That is, when getting on, it can be detected by the disappearance of the user from the user detection range, and when getting off, it can be detected by the appearance of the user in the user detection range.

Further, when the movement information of the user is acquired using only the in-car surveillance camera 20 provided in the car, the identification information of the individual user detected by the in-car surveillance camera 20 and the ID number are linked. , The appearance time when the user appears in the car and the disappearance time when the user disappears from the car are stored for each ID number. Then, the log output can be acquired by associating the departure floor and the arrival floor for each ID number from the boarding machine information and the time information.

As described above, the present invention sets and assigns discrimination information and ID information for each user to discriminate the user from the image of the boarding camera that shoots the landing of the elevator or the camera in the car that shoots the inside of the car. However, the movement of the user is traced based on the discrimination information and the ID information, and at least the boarding floor and the disembarking floor of the user are linked and stored to grasp the movement information of the user.

According to this, since the movement information of the user can be accurately grasped, for example, accurate parameter data necessary for prior simulation for performing group management control can be obtained, so that the accuracy of the simulation can be improved.

The present invention is not limited to the above-described embodiment, and includes various modifications. For example, the above-mentioned examples have been described in detail in order to explain the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the described configurations. Further, it is possible to replace one means of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. Further, it is possible to add, delete, or replace another configuration with respect to one means of the configuration of each embodiment.

10 ... Elevator operation management system, 10A ... Learning unit, 10B ... Reception unit, 10C ... Number of people evaluation unit by floor, 10D ... Comprehensive evaluation unit, 10E ... Allocation command unit, 11A-11N ... Boarding machine elevator control system, 12 ... Communication network, 16 ... Surveillance camera control system, 16A ... Image input processing unit, 16B ... Detection range setting processing unit, 16C ... User detection processing unit, 16D ... ID setting processing unit, 16E ... User detection processing unit by floor , 16F ... Movement information output processing unit, 16G ... Log output processing unit, 17 ... Landing monitoring camera, 18A-18D ... Landing door, 19A-19D ... User detection range, 20 ... Car monitoring camera.

Claims (6)

It is an elevator use log output system that analyzes the images of the landing cameras that take pictures of the landings of the boarding units of multiple elevators and grasps the movement information of the users of the landing.
The elevator use log output system is
A user detecting means for detecting discrimination information for discriminating an individual user from the image of the landing camera, and a user detecting means.
An ID number setting means for setting and assigning an individual ID number corresponding to the discrimination information,
A movement information detecting means for detecting the movement information of an individual user from the discrimination information and the ID number, and
A storage means for storing the movement information detected by the movement information detecting means corresponding to the ID number is provided .
Further, it is provided with a user detection range setting means for setting a user detection range for extracting a user for each boarding machine of the landing from the image of the landing camera.
The user detection means extracts the discrimination information of an individual user from the user detection range for each boarding machine set by the user detection range setting means.
Further, when a user exists in the overlapping area of the adjacent user detection range set by the user detection range setting means, the image of the user in the overlapping area is analyzed and the user exists in the overlapping area. It is equipped with a user detection range resetting means for setting the user detection range of the user, and also has a user detection range resetting means.
The user detection range resetting means sets the waiting direction of each boarding machine, detects the waiting direction of the user in the overlapping area from the image of the landing camera, and compares it with the waiting direction of the boarding machine. To set the user detection range of the users existing in the overlapping area.
Elevator usage log output system featuring this. The elevator use log output system according to claim 1.
The movement information detecting means is
The discrimination information at the time of boarding is compared with the discrimination information of an individual user who gets on the boarding machine from the boarding place and the discrimination information of an individual user who gets off from the boarding machine to another boarding point. Use of an elevator provided with a user recognition means for identifying individual users whose discrimination information matches when getting off, and detecting the movement information of individual users recognized by the user recognition means. Log output system. The elevator use log output system according to claim 2.
The elevator use log output system is
When the movement information is stored in the storage means, the ID number of each individual user and the ID number release means associated with the ID number are released according to a predetermined release condition.
Elevator usage log output system featuring this. This is an elevator usage log output method that analyzes the images of the landing cameras that capture the landings of the boarding units of multiple elevators and grasps the movement information of the users of the landing.
The elevator use log output method is
From the image of the landing camera, the discrimination information for discriminating an individual user is detected, and the discrimination information is detected.
Individual ID numbers are set according to the discrimination information, and
The movement information of an individual user is detected from the discrimination information and the ID number, and the movement information is detected.
The detected movement information is stored corresponding to the ID number, and the detected movement information is stored.
In addition, a user detection range for extracting users from the image of the landing camera for each boarding machine at the landing is set.
The discrimination information of an individual user is extracted from the user detection range for each set boarding machine.
When a user exists in an adjacent overlapping area of the user detection range, the image of the user in the overlapping area is analyzed and the user detection range of the user existing in the overlapping area is reset.
The waiting direction of each of the boarding units is set, and the waiting direction of the user in the overlapping area is detected from the image of the landing camera.
The user detection range of the user existing in the overlapping area is set as compared with the waiting direction of the boarding machine.
Elevator usage log output method characterized by this. The elevator usage log output method according to claim 4 .
When detecting the movement information,
The discrimination information of an individual user who gets on the boarding machine from the boarding point is compared with the discrimination information of an individual user who gets off from the boarding machine to another boarding point, and the discrimination information at the time of boarding and getting off are compared. The movement information is detected by identifying individual users who match the discrimination information at the time.
Elevator usage log output method characterized by this. The elevator usage log output method according to claim 5.
The elevator use log output method is
When the movement information is stored, the ID number of an individual user and the discrimination information associated with the ID number are released according to a predetermined release condition.
Elevator usage log output method characterized by this.

Images