JP4402505B2 - Waiting time guidance system and method - Google Patents

Description

The present invention relates to a waiting time guidance system and method for guiding waiting time required for a user to use a facility to be used, particularly in an ATM (Automatic Teller Machine) corner, an amusement park, a theme park, etc. of a financial institution. The present invention relates to a waiting time guidance system and method for accurately calculating the waiting time of a user who uses a facility to be used such as an ATM, a vehicle, and an attraction, and performing guidance.

  In general, when a user uses a facility such as an ATM at a financial institution ATM, amusement parks, theme parks, or other vehicles or attractions, the user can often queue in front of the facility. is there. In this case, the user who forms this queue waits for the turn to come, and can use the use target facility when the turn comes.

  However, in this type of facility target facility queue, there are cases where the queue arrangement of the target facility facility is restricted, such as the shape of the space in which the target facility queue can be queued, and the arrangement of the queue may be restricted by placing a rope or the like. A rope or the like that regulates how the queues are arranged may be complicatedly changed depending on the number of users who form the queue.

  For this reason, it is difficult for each user who forms a queue to determine how long the user can use the facility.

  Conventionally, systems disclosed in Patent Document 1 and Patent Document 2 are known as systems for guiding the waiting time to the user when the user forms a queue for using the target facility. ing.

  The prior art disclosed in Patent Document 1 can easily guide the waiting time of all the use facilities scattered in the wide entrance facility to the user, and can be used in a place where each use facility can be queued. A plurality of optical sensors that detect the presence of the customer are arranged along the lines, and the output of the plurality of optical sensors is checked in order from the optical sensor installed at the position closest to the entrance of the use facility. The queue length is comprehensively determined based on the combination of the output states and the cumulative results of the time series, the waiting time is obtained based on the queue length, and this waiting time is transmitted to the centralized display device. The waiting time of all the use facilities scattered in the wide entrance facility is centrally displayed.

In addition, the prior art disclosed in Patent Document 2 can accurately and automatically detect and notify the waiting time of a queue without requiring an attendant and without requiring a route by a rope or a simple fence. Specifically, by analyzing the image captured by the TV camera and recognizing a human aggregate, the largest area of the recognized aggregate is detected as a human matrix, The number of people in the matrix is obtained by dividing the area of the detected matrix by a predetermined unit area, and the waiting time is obtained by dividing the number of people by the number of matrix former visitors per unit time. Configured to detect.
JP 08-123374 A JP 2002-329196 A

  However, in the prior art disclosed in Patent Document 1, since the queue length is determined using an optical sensor and the waiting time is obtained based on the queue length, only a rough waiting time is displayed. As a result, the error between the actual time to reach and the display time increases, and the larger the difference between the expected waiting time and the actual waiting time, the greater the error. There is a problem of giving pleasure.

  Further, in the prior art disclosed in Patent Document 1, since a plurality of optical sensors are arranged in advance along a place where a queue can be made, for example, a rope or the like is stretched like an ATM corner of a financial institution. This method is not applicable in the case where the arrangement of queues is regulated and the rope or the like that regulates the arrangement of the queues is appropriately changed depending on the space, the number of users forming the queue, or the like.

  Further, in the prior art disclosed in Patent Document 2, the number of matrixes is obtained by dividing the area of the matrix by a predetermined unit area, and the number of matrix elements per unit time is determined in advance. Since the waiting time is detected by dividing by the number of visitors, it is impossible to accurately detect the waiting time of the customer, and it seems that the arrangement of the queue is regulated by using a rope etc. In such a case, there is a problem that the waiting time cannot be detected more accurately.

Accordingly, the present invention provides a latency guidance system and Methods capable of detecting and informing the exact user waiting time, even if as to regulate the arrangement of queues Put the rope The purpose is to do.

The present invention also aims to provide a time available notification intelligible latency to the user guidance system and Methods.

In order to achieve the above object, an invention according to claim 1 is a waiting time guidance system for guiding waiting time of a user who waits for use of at least one use target facility, and an imaging device that captures an image including the user. the comprising an image control apparatus for guiding the latency of the user based on the image captured by the imaging device, the image control apparatus, with the passage of time by framing the images captured by the imaging device Image framing means for storing a plurality of sheets, blocking means for dividing the image framed by the image framing means to create a block of a certain size, and the past of the blocks created by the blocking means a correlation coefficient calculating means for calculating a correlation coefficient between blocks each over a frame to the current, calculated by the correlation coefficient calculating means Labeling blocks the correlation coefficient is equal to or less than a predetermined value, recognizing means for recognizing a person label above a certain area, across the frame of the past to the present, the person recognized by said recognizing means a movement trajectories storage means for storing the motion as moving track, retrieve the moving track stored by said movement locus storage means, and calculates a variance of the moving track, the calculated issued variances portion below a certain value Partition line calculation means for calculating the partition line based on the display line, and display means for displaying a schematic image of the queue obtained by synthesizing the person recognized by the recognition means and combining the partition lines calculated by the partition line calculation means characterized by including and.

The invention of claim 2 is the invention of claim 1, the queue in which the user by the movement locus of the user who remembers by said movement locus storage means is formed by utilizing the usage target facility use and queuing identification means for identifying, said user the usage target facility based on the queue of the user identified in the mean time and the queue identification means required for use of the available target facility that has been set in advance Waiting time calculating means for calculating a waiting time until it becomes possible, and the display means displays the waiting time calculated by the waiting time calculating means together with a schematic image displayed by the display means. Features.

Further, the invention of claim 3, in the invention of claim 2, wherein the waiting time calculating means, the number of users of up to users that from the beginning of the user to form a queue identified by the wait queue identification means to a value obtained by multiplying the average usage time per person the user required to use the previous SL usage target facility quotient of the quotient and remainder value obtained by dividing the total usage target number of facilities of the utilization object facility, Ri the excess The waiting time until the certain user uses the use target facility is calculated by adding the current remaining use time of the use target facility having the short remaining time indicated by the value .

Further, the invention of claim 4, in the invention of claim 2 or 3, wherein the display means displays the waiting time of all users or specific users of the user forming the queue It is characterized by that.

According to a fifth aspect of the present invention, in the waiting time guidance method for guiding a waiting time of a user who waits for the use of at least one use target facility, an image including the user is picked up by an image pickup means, and the image pickup means the images captured by the plurality stored as image frames over time and frame by using the image frame means, creates a certain size of blocks each divided by the block catheter stage of said image frame, wherein With respect to the created block, the correlation coefficient between each block over the frame from the past to the present is calculated by the correlation coefficient calculation means, and the block where the correlation coefficient is equal to or less than a predetermined value is labeled, the label area is recognized as a person, the stored in the storage unit the movement of the past from the person who the recognized over the frame to the current as a locus, The cut line calculating means eject the movement trajectory that is the storage, dispersed calculates the movement locus, the calculated issued dispersion performs calculation of partition Ri line based on the following partial constant values, recognized A schematic image of the queue obtained by synthesizing a person and combining the partition lines detected by the partition line calculation unit is displayed on a display unit .

  According to the present invention, an image including a user who waits for the use target facility is captured, the image including the captured user is analyzed to detect a user's movement trajectory, and the detected user's movement The queue formed by the user is identified based on the trajectory, and the waiting time until the user can use the target facility based on the unit average time required for using the target facility and the identified user queue Is calculated, and the calculated waiting time is notified. For example, as in an ATM corner of a financial institution, the arrangement of queues is regulated by a rope or the like, and the arrangement of the queues is regulated. It is possible to accurately detect and notify the user's waiting time even when the rope to be changed is appropriately changed depending on the space, the number of users forming the queue, etc. User understands It is possible to notify the waiting time in real time, as well as to notify the waiting time in real time and the waiting time for each user who forms a queue, and how long each user uses the target facility. There is an effect that it is possible to know exactly whether it is possible.

Hereinafter, an embodiment of a waiting time guidance system and method according to the present invention will be described in detail with reference to the accompanying drawings.

Note that the waiting time guidance system and method of the present invention is based on the waiting time of a facility to be used where a user forms a queue such as an ATM corner of a financial institution, an amusement park, a theme park or the like, a vehicle, or an attraction. Although applicable to notification, in the following embodiment, a case where the use target facility is an ATM corner of a financial institution will be described as an example.

  In the example shown in Example 1, the example in the case of using ATM as a utilization object facility is shown.

Figure 1 is a block diagram showing the system configuration of a waiting time during guidance system configured by applying latency guidance system and method according to the present invention.

  In FIG. 1, this waiting time guidance system includes an imaging unit 101, an identification unit 102, a queue detection unit 103, a waiting time calculation unit 104, a use average time calculation unit 105, a storage unit 106, an image image generation unit 107, and a notification unit. 108, a display unit 109, a broadcast unit 110, a partition line detection unit 111, a user detection unit 112, and a locus detection unit 113.

  The image capturing unit 101 captures an image of a user who is waiting for the use target facility.

  For example, if the facility to be used is an ATM (Automatic Teller Machine) of a financial institution, a wide-angle camera or the like is installed on the ceiling of the financial institution to wait for the use of the ATM. The user is imaged.

  The identification unit 102 identifies a user who is waiting in turn for using the ATM imaged by the imaging unit 101. Since the person is identified, the number of persons forming a queue (waiting in turn) can also be calculated.

  That is, the image picked up by the image pickup unit 101 includes an ATM that is a target facility other than the user, a rope for partitioning the user's procession, and other installations. And the number of identified persons can be calculated.

  The trajectory detection unit 113 detects a trajectory of a user who is waiting in order to use the ATM identified by the identification unit 102.

  The queue detection unit 103 detects that the user identified by the identification unit 102 via the trajectory detection unit 113 forms a queue. Here, the order for each user until the ATM is used in the queue formed by the user is also detected.

  The waiting time calculation unit 104 calculates a waiting time until the ATM is used.

  For calculating the waiting time, the average usage time per person calculated in advance by the average usage time calculation unit 105, the number of ATMs to be used, and the order in which the user uses the ATM in the queue detection unit 103 To calculate the queue.

  For example, the waiting time of each person in the matrix can be calculated by the following calculation. At this time, the following four constants are determined.

Number of processing units: m Average usage time per processing unit: A hours Elapsed usage time per processing unit: P1, P2, ..., P m
Remaining usage time from the average usage time per processor: R1, R2, ..., R m
Therefore, the remaining usage time of each processing device is
R1 = A-P1
R2 = A-P2
~
R m = A-P m .

  The processing device with the shortest remaining time can be calculated using the above constants.

  As this calculation method, it may be calculated by sorting remaining time, or a function for obtaining the minimum may be used. Further, the processing device having the shortest remaining time may be selected as the processing device having the shortest remaining time among the processing devices remaining without the processing device having the shortest remaining time. By repeating this, it is possible to calculate the order of the remaining usage time of each processing device at a certain point in time, and to calculate the waiting time for each person at a certain point in time.

In general, the waiting time Wn of the nth person can be expressed as follows, where the remainder of “n ÷ m ” is “s” and the quotient is “t”.

『Wn = Rs + A * t』
“Rs” at this time is the remaining use time of the processing apparatus with the shortest remaining time.
Thereby, the waiting time of each person forming the queue can be calculated.

  Of course, the waiting time of an arbitrary person may be calculated using queuing theory (model).

  The average usage time calculation unit 105 calculates how much time the ATM, which is a usage target facility, is averaged per person and used.

  In other words, the average time of users used in the past may be calculated and set in advance, or the latest average usage time may be calculated at all times and the data may be used as the average usage time.

  The storage unit 106 stores the average usage time per person calculated by the average usage time calculation unit 105.

  The image image generation unit 107 synthesizes an image schematically obtained from the image captured by the image capturing unit 101 with respect to the person identified by the identification unit 102 and the image schematically representing the partition line detected by the partition line detection unit 111. Thus, a simulated image that closely resembles an actual image captured by the imaging unit 101 is generated.

  At this time, as the image data transmitted from the queue detection unit 103, the actual image captured by the imaging unit 101 and the number of persons identified by the identification unit 102 are also transmitted at the same time as the data indicating the detection of the queue. Therefore, the image image generation unit 107 can generate a schematic image that closely resembles an actual image.

  The notification unit 108 includes a display unit 109 and a broadcast unit 110, and is calculated by the waiting time calculation unit 104 in addition to the schematic image image generated by the image image generation unit 107 for the users who form a queue. This waiting time is notified to all users or specific users among the users.

  The display unit 109 calculates a waiting time in addition to an image that closely resembles an actual image obtained by combining the image image generated by the image image generation unit 107 and the partition line detected by the partition line detection unit 111. The waiting time for all the users calculated by the unit 104 or for a specific user can be displayed on the display at the same time.

  Further, since the number of persons identified by the identification unit 102 is sent to the display unit 109 via the queue detection unit 103 and the image image generation unit 107, the display content (schematic image image shown above) is displayed. In addition to the waiting time), the waiting number can also be displayed.

  The broadcast unit 110 notifies the queue state by voice.

  The partition line calculation unit 111 calculates a partition line based on the movement trajectory of the person identified by the identification unit 102.

  The user detection unit 112 detects whether or not the user has moved across the partition line detected by the partition line detection unit 111 based on the trajectory information of the person identified by the identification unit 102, and the person who has moved across the partition line Is detected, a request is made to display a person who has moved across the partition line as an image different from that of other special persons among the images generated by the image image generation unit 107.

  For example, by blinking a schematic image of a person who has moved across a partition line, a different display can be obtained.

  At this time, a specific person such as an administrator or employee is configured not to display an image by holding an IC (Integrated Circuit) chip or the like, or straddles a partition line. It may be configured to display without blinking even when moved by.

  According to such a configuration, the person imaged by the imaging unit 101 can be uniquely identified by the identification unit 102, so that the number of people forming the matrix can be calculated, and all of the waiting time calculation unit 104 can The waiting time for each user or specific user can be calculated. Then, based on the image image of the person captured by the image capturing unit 101 and the partition line detected by the partition line detection unit 111 based on the trajectory detected by the trajectory detection unit 113, the image is calculated. The waiting time and the waiting number can be displayed together.

  This makes it possible to display the state of the queue in consideration of privacy, and to display the optimal image because it automatically calculates without setting the position of the dividing line that changes every day. In addition, it is possible to display an appropriate waiting time for each user in an easy-to-understand manner.

  FIG. 2 is a layout diagram of the waiting time guidance system according to the present invention.

  FIG. 2 shows an example in which image control of persons lined up at an ATM of a financial institution is performed. FIG. 2 shows ATM (201-A, 201-B, 201-C), ATM user (202-A, 202-B, 202-C), waiting person 203, rope 204, wide angle camera 205, image control. A device 206 and a display 207 are provided.

  The ATM user (202-A, 202-B, 202-C) operates the ATM (201-A, 201-B, 201-C), and the waiting person 203 is the ATM user (202-A, 202-C). B, 202-C) waits within the rope 204 for the operation to end.

  The wide-angle camera 205 monitors and detects the status of the ATM user (202-A, 202-B, 202-C) and the waiting person 203, and sends the detected person to the image control device 206. Based on the transmitted data, the image control device 206 generates an image image close to the actual image, detects a partition line, calculates the waiting time waiting number, and displays it on the display display 207 together with the image image. .

  FIG. 3 is a diagram showing an image displayed on the display according to the present invention.

  In FIG. 3, ATM 301, ATM user 302, waiting person 303, partition line 304, current waiting time 305, current waiting number 306, remaining waiting time of waiting person (307-A, 307-B) are shown. it's shown.

  The image displayed on the display is an image close to the actual image. By combining the various display data results calculated by the method shown in FIG. 1 with this, the partition line 304, the current waiting time 305, the current waiting number 306, and the remaining waiting time (307-A, 307-B) are calculated. Will be able to.

  FIG. 4 is a flowchart showing an outline of the entire processing of the waiting time guidance system according to the present invention.

  In FIG. 4, the process is started from shooting a person forming a queue (step 401), and information on a movement trajectory indicating the movement of the photographed person (hereinafter referred to as trajectory information) is generated. (Step 402). The generated movement trajectory information is stored (step 403), and an image image close to a person photographed in parallel with the generation of a partition line image from the stored trajectory information (step 404) is generated (step 405). .

  The captured image and the waiting number data of the person are calculated (step 406), and these are displayed on the display (step 407).

  FIG. 5 is a view showing a recognized object taken by the wide-angle camera of the waiting time guidance system according to the present invention.

  In FIG. 5, ATM (501-A, 501-B, 501-C), ATM user 502, waiting person 503, and rope 504 are displayed.

  The ATM (501-A, 501-B, 501-C), the ATM user 502, and the waiting person 503 are recognized by the image captured by the wide-angle camera, and the rope 504 is determined from the captured image. In view of the fact that the rope 504 cannot be detected because the set position of the rope varies from day to day, it is indicated by a dotted line.

  FIG. 6 is a flowchart showing person recognition and trajectory information creation processing of the waiting time guidance system according to the present invention.

  In FIG. 6, when an image is input by a wide-angle camera (step 601), the processing is started, and the input image is divided into a certain size (hereinafter referred to as a block) (step 602), and each of the divided blocks is divided. In step 603, correlation coefficients of the past several frames are calculated. Next, a block having a certain threshold value or less is detected from the calculated correlation coefficients (step 604). A labeling process for classifying a plurality of regions as a group by adding the same label to the pixels connected in the detected blocks is executed (step 605). This makes it possible to count how many independent regions are in the image (labeling) and to examine the shape of the labeled blocks.

  It is determined whether the area of the shape of the labeled block is a predetermined area or more (step 606). If the area is less than the predetermined area (NO in step 606), the process ends without recognizing the person. To do. On the other hand, if the object has a certain area or more (YES in step 606), it is recognized as a person (step 607), the optical flow of the person is calculated from the past several frames, and the trajectory for each person is calculated. Is calculated (step 608).

  It is determined whether or not the calculated trajectory information for each person already exists (step 609). If a trajectory for the person already exists (YES in step 609), the trajectory existing for the person is extended (step 610). . Further, for a person for which there is no locus calculated for each person (NO in step 609), new locus information is generated as a person newly added to the matrix (step 611).

  As a result, only the person is recognized, and the trajectory information for the person can be generated.

  FIG. 7 is a diagram showing a calculation image of the partition lines in the present invention.

  FIG. 7 shows a state in which trajectory information when a person passes is shown. In the example shown in FIG. 7, trajectory information of five persons (A to E) is shown.

  When the trajectory information on the persons (A, B, C) receiving the service is confirmed, it can be confirmed that the trajectory information of the persons A and C is widely dispersed (701). When the trajectory information before the service is received is confirmed, it can be recognized that the variance of the trajectory information is small (702, 703). Then, when the trajectory information is traced from the ATM device 700 providing the trajectory information, there is a point where the variance of the trajectory information increases (704).

  Considering the variance of the trajectory information, it can be divided into variance suppressed to a certain value and variance above a certain value, and the trajectory information part that forms variance below a certain value moves within the partition. Can be judged. Thereby, the partition line can be calculated on the basis of the locus information of the dispersed portion having a certain value or less.

  FIG. 8 is a flowchart showing a partition line calculation process according to the present invention.

  In FIG. 8, processing is started when the trajectory information updated within a certain time period for each person already stored (step 801), and the process is started. Among the extracted trajectory information, the variance value is equal to or smaller than a certain value. It is determined whether there is a part (step 802), and if there is no part below a certain value (NO in step 802), it is determined that a matrix is not formed or a service is not provided and processing is performed. finish. On the other hand, when there is a portion where the variance value is equal to or less than a certain value (step 802), a partition line is calculated in the portion where the variance value is equal to or less than the certain value (step 803).

  Next, it is determined whether or not the existing partition line data exists around the trajectory information below a certain value (step 804). If there is no existing partition line data (NO in step 804), the partition line calculated by the above processing is displayed. (Step 805). If a partition line exists (YES in step 804), a comparison is made as to whether the partition line data that has existed before matches the new partition calculated in step 803 (step 806). If YES (YES in step 806), it is determined that an existing partition is to be used, and the process ends. If they do not match (NO in step 806), the partition line data calculated in step 803 is stored as new partition line data (step 807).

  Furthermore, it is determined whether or not the candidate matches the previous number of frames (step 808). If the candidate does not match (NO in step 808), the process ends. If they match (YES in step 808), the stored partition is displayed as a new partition line (step 809).

  In addition, the display content on the display when there is no trajectory information is to display a default screen or a content that indicates that no service is formed and a service can be provided immediately. Alternatively, nothing may be displayed.

  FIG. 9 is a diagram illustrating an image when a movement across the calculated partition line occurs.

  FIG. 9 shows the calculated partition line 901, the person 902 who has moved across the partition line, and the trajectory information 903 traced by the person who has moved across the partition line.

  When the person 902 who has moved across the partition line moves from the outside to the inside of the partition line, or moves from the inside to the outside, it is determined that the person 902 has crossed the partition line. "When moving from the inside to the outside, it is assumed that the user has left the matrix." That is, when the person moves from the outside to the inside, the person is determined to be an interrupting person, and the movement from the inside to the outside is determined to straddle the partition line, but is not determined to be “interrupt” but is determined to be “out of the matrix”.

  Here, the inside means the side where the trajectory information used when generating the partition line is present, and the outside means a portion that is not inside. Further, there is an effect that an interruption can be prevented by making the display format of the person 902 moved across the partition different from other persons.

  Further, a person who straddles the partition line may hold a specific device such as an IC chip so as not to detect that the person straddles the partition line. Thereby, workability | operativity of a staff member or a specific worker can be improved.

  FIG. 10 is a diagram illustrating a warning method when a movement across a partition line occurs.

  FIG. 10 shows a warning screen 1001 for displaying on the monitoring screen that a movement has occurred across the partition line as a type of warning, an email 1002 when not pursuing real-time characteristics, and a warning lamp 1003. It is shown.

  For example, the warning screen 1001 is displayed on a monitoring monitor such as a centralized management office to notify the administrator, and the administrator who has received the notification can arrange the site.

  Further, it becomes possible to notify a staff member who is remote by e-mail 1002 or message transmission, and further, by providing a warning lamp 1003, interruption can be suppressed.

  FIG. 11 is a flowchart of processing for specifying a position for displaying a person.

  In FIG. 11, the trajectory information updated within a predetermined time is extracted from the stored trajectory information (step 1101), and the image of the person is displayed at the end point (end point) in the extracted trajectory information (step 1102). ).

  FIG. 12 is a flowchart of processing for displaying a person straddling a partition line.

  In FIG. 12, processing is started when the stored trajectory information is extracted (step 1201), and it is determined whether or not there is a trajectory straddling the partition line in the extracted trajectory information (step 1202). If there is no trajectory straddled (NO in step 1202), the process ends. If there is a trajectory straddling the partition line (YES in step 1202), among the persons having the trajectory straddling the partition, a person who has moved from the outside of the partition line to the inside is identified as an interrupter (step 1203). . For the person identified as the interrupter, a special person image is displayed on the display (step 1204).

  FIG. 13 is a flowchart of processing up to obtaining the waiting time waiting number.

  In FIG. 13, since it can be determined that a matrix is formed when there is at least one person inside the created partition line, it is first determined whether a person exists inside the partition line (step 1301). ), If no person is present (NO in step 1301), it can be determined that there is no matrix, and the process ends. If there is at least one person (YES in step 1301), the number of persons is counted by identifying each person who has stopped (in a matrix inside the partition line) (step 1302).

  The waiting time is calculated for each identified person using the method described in FIG. 1 (step 1303), and the waiting time is displayed for each counted person (step 1304).

  Through the above processing, the waiting time guidance system of the present invention can recognize a queue of people waiting for service provision, and can display the matrix of the recognized persons with image images close to real images. A partition line, such as a rope, that forms the path that forms the path can be automatically generated from the trajectory information of the person and displayed together with the image. In addition, it is possible to display a more accurate waiting time and waiting number of people until the service is provided by the queue, and it is also possible to display the waiting time of an arbitrary person in the queue.

  FIG. 15 is a diagram illustrating a method using a standard deviation based on a person's movement trajectory as a calculation method when displaying a partition line on the screen.

  A standard deviation is calculated in consideration of variance based on a movement trajectory formed by a plurality of persons, and the movement trajectory at the calculated standard deviation is a curve 1501 (hereinafter referred to as “end point”) shown in FIG. Standard curve). Using the distance AB from the preset standard curve 1501, two curves 1502 separated by a new distance AB (hereinafter referred to as AB standard curve) are generated on both sides of the standard curve (dotted curve in FIG. 15). .

  At this time, a two-dimensional plane which is an X, Y plane is considered as shown in FIG. A normal vector 1503 of the two generated AB standard curves 1502 is obtained for each coordinate, and an intersection 1504 between normal vectors having the maximum inclination angle of the normal vector 1503 (that is, orthogonal to the X and Y axes) is obtained. calculate. Of the intersections 1504 of the calculated normal vectors, intersections that do not intersect the standard curve are connected by a straight line to obtain a partition line. When the angle of the normal vector 1503 at the adjacent point becomes maximum, only the normal vector according to the set content is used based on information set in advance.

  In addition to this, it is assumed that various settings are made in advance for calculating the optimum partition line.

  FIG. 16 is a diagram in which the partition lines and the standard curve calculated in the second embodiment of the present invention are two-dimensionally displayed.

  In FIG. 16, the standard curve 1601 and the partition line 1602 calculated in FIG. 15 are displayed, and the partition line of the three-dimensional space is displayed on the display display using the calculated partition line.

  As a result, it is possible to show an image close to the real image of the queue to the person waiting for the turn, and to accurately convey the time and number of people until the service is provided.

  Therefore, by applying the present invention, it becomes possible to display a matrix image that is easy to understand for the user, and further, it is possible to protect privacy by using an image close to the actual image. I can expect. Moreover, the effect that the position of the rope converted according to the day can be input in advance or the image of the partition can be displayed without setting can be expected.

  The present invention is not limited to the embodiments described above and shown in the drawings, and can be implemented by being appropriately modified within the scope not changing the gist thereof.

  For example, the above embodiment shows an example in which the service providing device is an ATM, but it can also be realized in a scene where a plurality of persons such as attractions in an amusement park or a ticket office form a queue. It may be configured.

  Further, in the first embodiment, the queue image diagram and the waiting time are displayed at the same time, but each may be displayed separately. For example, as shown in FIG. 14, a screen 1400 displaying information on only “current waiting time” and “current waiting number of people” is arranged at the end of the queue or near the entrance so that information necessary for the user can be quickly obtained. Can communicate accurately.

  In addition, if a certain value is exceeded according to the calculated waiting time or the number of waiting people, the usage fee is discounted, a fee is set for a user who uses the user for a long time, or the fee is increased. A built-in configuration may be used.

  INDUSTRIAL APPLICABILITY The present invention is applicable to a waiting time guidance system that displays the state of a queue formed by a user who waits for the use of a facility to be used, and in particular, the state of the formed queue is schematically illustrated to be close to a real image. An image is generated and displayed with the partition lines detected from the trajectory information of the person, and moreover, the waiting time until the user can use the target facility and the number of people waiting are displayed more accurately. Useful for displaying the waiting time for each.

The block diagram which shows the system configuration | structure of the waiting time guidance system concerning this invention. 1 is a layout diagram of a waiting time guidance system according to the present invention. The figure which shows the image displayed on the display display concerning this invention. The flowchart which shows the outline of the whole process of the waiting time guidance system concerning this invention. The figure which shows the recognition thing of the image image | photographed with the wide angle camera of the waiting time guidance system concerning this invention. The flowchart which shows the person recognition of the waiting time guidance system concerning this invention, and the creation process of locus information. The figure which shows the calculation image of the partition line in this invention. The flowchart which shows the calculation process of the partition line in this invention. The figure which shows an image when the movement across the calculated partition line in this invention generate | occur | produces. The figure which shows the method of warning when the movement which straddles the partition line in this invention generate | occur | produces. The flowchart of the process which specifies the position which displays the person in this invention. The flowchart of the process which displays the person who straddled the partition line in this invention. The flowchart of a process until it calculates | requires waiting time waiting number in this invention. The figure which shows the waiting time waiting number at the time of dividing the display of the image image and waiting time waiting number in this invention. The figure which shows the calculation method of the partition line in the other Example of the waiting time guidance system concerning this invention. The figure which displayed the partition line and standard curve which were calculated in the other Example in this invention two-dimensionally.

Explanation of symbols

DESCRIPTION OF SYMBOLS 101 Image pick-up part 102 Identification part 103 Trajectory detection part 104 Wait time calculation part 105 Use average time calculation part 106 Storage part 107 Image image generation part 108 Notification part 109 Display part 110 Broadcasting part 111 Partition line detection part 112 User detection part 113 Trajectory Detection unit 201-A, 201-B, 201-C ATM
202-A, 202-B, 202-C ATM user 203 Waiting person 204 Rope 205 Wide-angle camera 206 Image controller 207 Display display

Claims (5)

In a waiting time guidance system that guides the waiting time of a user waiting for the use of at least one target facility,
An imaging device that captures an image including the user;
An image control device for guiding the waiting time of the user based on an image captured by the imaging device ;
Comprising
The image control device includes:
Image framing means for framing an image captured by the imaging device and storing a plurality of images over time ;
And blocking means for creating a certain size of blocks by dividing a picture that has been framed by the image framing means,
Correlation coefficient calculating means for calculating a correlation coefficient between the blocks over the frame from the past to the present of the blocks created by the blocking means;
Blocks the correlation coefficient calculated by the correlation coefficient calculating means is equal to or less than a predetermined value and labeling, recognizing means for recognizing a person label above a certain area,
Movement trajectory storage means for storing the movement of the person recognized by the recognition means as a movement trajectory over the frame from the past to the present ;
Removed movement locus stored by said movement locus storage means, distributed calculates the movement trajectory, and a partition line calculation means the calculated issued variances to calculate the partition line on the basis of the following parts a certain value ,
A waiting time guidance system comprising: a display unit configured to display a schematic image of the queue obtained by synthesizing the partition line calculated by the partition line calculation unit and the person recognized by the recognition unit . Queue identifying means for identifying a queue formed by the user until the user uses the facility to be used, based on the movement locus of the user stored by the movement locus storage means ;
To the user based on the queue of identification by said user in the mean time and the queue identification means required for the use of preset the usage target facility is available for the use target facility latency Waiting time calculating means for calculating
Comprising
The display means includes
The waiting time guide system according to claim 1, wherein the waiting time calculated by the waiting time calculation means is displayed together with a schematic image displayed by the display means . The waiting time calculating means includes
Wherein the quotient of the queue identification means all available target facility number in the quotient and remainder values of the head up to the user that the user number of users the utilization object property to form a the queue identified by to use a value obtained by multiplying the average usage time per person the user required for use of the target facility, in Rukoto be added to the current usage time remaining usage target facilities remaining time in which the remainder value shown is in short order, The waiting time guidance system according to claim 2, wherein a waiting time until the certain user uses the facility to be used is calculated . The display means includes
Latency guidance system according to claim 2 or 3, characterized in that displaying the waiting time of all users or specific users of the user forming the queue. In the waiting time guidance method for guiding the waiting time of a user waiting for the use of at least one target facility,
Taking an image including the user with an imaging means,
The image captured by the imaging unit is framed using an image framing unit, and a plurality of images are stored as image frames over time ,
Create a certain size of blocks respectively divided by the blocking means of said image frame,
With respect to the created block, the correlation coefficient between each block over the frame from the past to the present is calculated by the correlation coefficient calculation means,
Labeling the block whose correlation coefficient is a predetermined value or less , recognizing a label having a certain area or more as a person,
The movement of the recognized person over the frame from the past to the present is stored in the storage unit as a trajectory ,
The stored movement trajectory is taken out by the partition line calculating means, the variance of the movement trajectory is calculated, and the partition line is calculated based on a portion where the calculated variance is a predetermined value or less.
A waiting time guidance method , wherein the recognized person is modeled, and a schematic image of the queue obtained by combining the partition lines detected by the partition line calculation means is displayed on the display means .

Images