HK1238230B - Riding detection system of elevator - Google Patents

Description

Elevator riding detection system

This application is based on and claims the benefit of priority from Japanese Patent Application No. 2016-4593 (filing date: January 13, 2016), the entire contents of which are incorporated herein by reference.

Technical Field

An embodiment of the present invention relates to an elevator boarding detection system for detecting a user boarding a passenger car.

Background Art

Typically, when an elevator car arrives at the elevator lobby and opens its doors, it closes and departs after a predetermined time. Elevator users are unaware of the closing time of the car door and may bump into the closing car door when entering the elevator lobby.

To prevent such door collisions during boarding, one approach is to use sensors to detect passengers entering the car and control the door's opening and closing. A photoelectric sensor is typically used as such. Specifically, a photoelectric sensor is installed above the car to optically detect passengers entering the car. While a passenger is detected, the door remains open, preventing the passenger from bumping into the closing door and being pulled into the door frame.

Summary of the Invention

However, photoelectric sensors have a narrow detection range and can only detect users by pinpoint. Consequently, there are cases where a user is slightly away from the car and the door closes without being detected, or vice versa, the door opens due to the mistaken detection of a person passing by the car.

The problem to be solved by the present invention is to provide an elevator boarding detection system that can accurately detect users intending to board an elevator over a wide range and reflect the result in the door opening and closing control.

An elevator occupancy detection system in one embodiment includes: a camera unit, which can capture a specified range from near the door of the passenger car toward the elevator lobby when the passenger car has arrived at the elevator lobby; a user detection unit, which uses a plurality of images taken by the camera unit in a time series to focus on the activities of people/objects in a pre-set area to detect the presence of users who intend to board the elevator; and a control unit, which controls the opening and closing of the door based on the detection result of the user detection unit.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a diagram showing the configuration of an elevator riding detection system according to one embodiment.

FIG. 2 is a diagram showing an example of an image captured by a camera in this embodiment.

FIG. 3 is a diagram showing a state in which a captured image is divided into blocks in the embodiment.

FIG. 4 is a diagram for explaining a detection area in a real space in this embodiment.

FIG. 5 is a diagram for explaining a coordinate system in a real space in this embodiment.

FIG. 6 is a diagram for explaining motion detection using image comparison in this embodiment, and schematically shows a portion of an image captured at time t.

FIG. 7 is a diagram for explaining motion detection using image comparison in this embodiment, and schematically shows a portion of an image captured at time t+1.

FIG8 is a flowchart showing the overall processing flow of the riding detection system in this embodiment.

FIG9 is a flowchart showing the motion detection process of the riding detection system in this embodiment.

FIG10 is a flowchart showing the position estimation process of the riding detection system in this embodiment.

FIG. 11 is a flowchart showing a riding intention estimation process of the riding detection system in this embodiment.

FIG. 12 is a diagram showing a change in the foot position of a person intending to ride in the embodiment.

FIG. 13 is a diagram showing a change in the foot position when the rider has no intention to ride in the embodiment.

FIG. 14 is a diagram for explaining motion detection using region segmentation of a captured image in another embodiment.

FIG. 15 is a diagram for explaining a method for dealing with a case where a camera is installed at an angle in another embodiment.

FIG. 16 is a diagram for explaining another method of coping with a situation in which a camera is tilted in another embodiment.

FIG17 is a flowchart showing the processing operation of the riding detection system in another embodiment during the closing of the passenger car door.

DETAILED DESCRIPTION

Hereinafter, embodiments will be described with reference to the accompanying drawings.

Fig. 1 is a diagram showing the configuration of an elevator passenger detection system according to one embodiment. Note that, although a single passenger car is used as an example for explanation, a plurality of passenger cars also have the same configuration.

A camera 12 is installed above the entrance and exit of the passenger car 11. Specifically, the lens portion of the camera 12 is installed on the lintel 11a covering the upper entrance and exit of the passenger car 11, facing the elevator lobby 15. The camera 12 is a small surveillance camera, such as an on-board camera, equipped with a wide-angle lens, capable of continuously capturing several frames per second (e.g., 30 frames per second). When the passenger car 11 reaches each floor and opens its door, the camera captures the state of the elevator lobby 15, including the state near the car door 13 inside the passenger car 11.

The imaging range at this time is adjusted to L1+L2 (L1>>L2). L1 is the imaging range on the elevator lobby side, which is, for example, 3 meters from the car door 13 toward the elevator lobby 15. L2 is the imaging range on the car side, which is, for example, 50 cm from the car door 13 toward the rear of the car. Furthermore, L1 and L2 are the ranges in the depth direction, and the range in the width direction (a direction perpendicular to the depth direction) is at least larger than the horizontal width of the passenger car 11.

Furthermore, in the elevator lobby 15 on each floor, a hall door 14 is provided at the entrance to the passenger car 11, allowing for flexible opening and closing. The hall door 14 engages with the car door 13 upon arrival of the passenger car 11, opening and closing. The power source (door motor) is located on the passenger car 11 side, and the hall door 14 opens and closes simply following the car door 13. In the following description, it is assumed that the hall door 14 opens when the car door 13 opens, and closes when the car door 13 closes.

Each image (video) captured by the camera 12 is analyzed and processed in real time by the image processing device 20. In FIG1 , the image processing device 20 is shown removed from the passenger car 11 for ease of explanation. However, in reality, the image processing device 20 is housed in the door lintel 11a together with the camera 12.

Here, the image processing device 20 is equipped with a storage unit 21 and a user detection unit 22. The storage unit 21 includes a buffer for sequentially storing images captured by the camera 12 and temporarily retaining data required for processing by the user detection unit 22. The user detection unit 22 detects the presence of users who intend to board the car by focusing on the movements of people/objects closest to the car door 13 in a plurality of time-sequential images captured by the camera 12. The user detection unit 22 is functionally divided into an activity detection unit 22a, a position estimation unit 22b, and a boarding intention estimation unit 22c.

The motion detection unit 22a compares the brightness of each image in units of blocks to detect the motion of people or objects. The "motion of people or objects" mentioned here refers to the motion of people, wheelchairs, and other moving objects in the elevator lobby 15.

The position estimating unit 22b extracts the block closest to the car door 13 from the blocks with motion detected by the motion detecting unit 22a for each image, and estimates the coordinate position (Y coordinate shown in FIG5 ) from the center of the car door 13 (the center of the door front width) in this block in the direction of the elevator lobby as the user's position (foot position). The boarding intention estimating unit 22c determines whether the user intends to board based on the time-series changes in the position estimated by the position estimating unit 22b.

It should be noted that these functions (the motion detection unit 22 a , the position estimation unit 22 b , and the boarding intention estimation unit 22 c ) may be provided in the camera 12 or in the car control device 30 .

The car control device 30 is connected to an elevator control device (not shown) and transmits and receives various signals, such as hall calls and car calls, to the elevator control device. An "elevator hall call" is a call signal registered by operating a hall call button (not shown) located in the elevator hall 15 on each floor, and includes information about the registered floor and destination direction. A "car call" is a call signal registered by operating a destination call button (not shown) located in the cabin of the passenger car 11, and includes information about the destination floor.

The car control device 30 also includes a door opening and closing control unit 31. The door opening and closing control unit 31 controls the opening and closing of the car door 13 when the passenger car 11 arrives at the elevator lobby 15. Specifically, the door opening and closing control unit 31 opens the car door 13 when the passenger car 11 arrives at the elevator lobby 15 and closes the car door 13 after a predetermined time has elapsed. However, if the user detection unit 22 of the image processing device 20 detects a person intending to board the car door 13 while the car door 13 is open, the door opening and closing control unit 31 prohibits the closing of the car door 13 and maintains the door open.

Next, the riding intention detection method in this embodiment will be described with reference to FIG. 2 to FIG. 7 .

Figure 2 shows an example of an image captured by camera 12. In the figure, E1 represents the estimated position area, and yn represents the Y coordinate of the detected position of the user's foot. Figure 3 shows the captured image after it has been divided into blocks. The portion of the original image divided into a grid with a side of W _blocks is referred to as a "block."

The camera 12 is installed above the entrance and exit of the car 11. Therefore, when the car 11 opens its door at the elevator lobby 15, it captures a predetermined range (L1) on the elevator lobby side and a predetermined range (L2) inside the car. Using the camera 12 expands the detection range, allowing detection of users who are slightly further away from the car 11. However, there is a risk of erroneous detection of a person simply passing through the elevator lobby 15 (not boarding the car 11) opening the car door 13.

Therefore, the present system is configured as follows: as shown in FIG3 , the image captured by the camera 12 is divided into blocks of a certain size, blocks with human/object activity are detected and tracked, thereby determining whether the user is willing to ride.

Furthermore, in the example of FIG3 , the blocks have the same length in both the vertical and horizontal directions, but the lengths in both directions may be different. Furthermore, the blocks may be uniformly sized across the entire image area, or may be non-uniformly sized, such as with the vertical (Y-direction) length decreasing toward the top of the image. This allows the foot position, which is subsequently estimated, to be determined at a higher resolution or with a uniform resolution in real space (if the image is uniformly divided, the resolution decreases with distance from the car door 13 in real space).

Fig. 4 is a diagram for explaining a detection area in a real space. Fig. 5 is a diagram for explaining a coordinate system in a real space.

In order to detect the activities of users who are willing to ride based on the captured images, first, an activity detection area is set for each block. Specifically, as shown in Figure 4, at least a position estimation area E1 and a riding intention estimation area E2 are set. The position estimation area E1 is an area for estimating a part of the body of the user walking from the elevator lobby 15 toward the car door 13, specifically, the position of the user's feet. The riding intention estimation area E2 is an area for estimating whether the user detected in the position estimation area E1 has the intention to ride. Furthermore, the riding intention estimation area E2 is included in the position estimation area E1 and is also an area for estimating the position of the user's feet. That is, in the riding intention estimation area E2, the position of the user's feet is estimated and the user's willingness to ride is estimated.

In real space, the position estimation area E1 is a distance L3 from the center of the car door 13 toward the elevator lobby, which is set to, for example, 2 meters (L3 ≤ the imaging range L1 on the elevator lobby side). The horizontal width W1 of the position estimation area E1 is set to a distance greater than the horizontal width W0 of the car door 13. The boarding intention estimation area E2 is a distance L4 from the center of the car door 13 toward the elevator lobby, which is set to, for example, 1 meter (L4 ≤ L3). The horizontal width W2 of the boarding intention estimation area E2 is set to a distance approximately the same as the horizontal width W0 of the car door 13.

Furthermore, the lateral width W2 of the boarding intention estimation area E2 may be larger than W0. Furthermore, the boarding intention estimation area E2 may not be rectangular in the real space but may be trapezoidal excluding the blind spot of the door frame.

Here, as shown in FIG5 , the camera 12 captures images with the X-axis being horizontal to the car door 13 provided at the entrance and exit of the passenger car 11, the Y-axis being the direction from the center of the car door 13 to the elevator lobby 15 (a direction perpendicular to the car door 13), and the Z-axis being the height of the passenger car 11. In each image captured by the camera 12, the position estimation area E1 and the boarding intention estimation area E2 shown in FIG4 are compared on a block basis, thereby detecting changes in the foot position of a user moving in the direction from the center of the car door 13 to the elevator lobby 15, i.e., the Y-axis direction.

This situation is shown in FIG6 and FIG7.

6 and 7 are diagrams for explaining motion detection using image comparison: FIG6 schematically shows a portion of an image captured at time tn, and FIG7 schematically shows a portion of an image captured at time tn+1.

P1 and P2 in the figure represent the portions of the captured image where user activity is detected. These are actually a collection of blocks where activity is detected through image comparison. The presence of a passenger's boarding intention is determined by extracting the block Bx closest to the car door 13 in image portions P1 and P2 and tracking the Y coordinate of this block Bx. In this case, if equidistant lines (equally spaced horizontal lines parallel to the car door 13) are drawn along the Y axis, as indicated by dashed lines, the distance between block Bx and the car door 13 in the Y axis direction can be determined.

In the examples of FIG. 6 and FIG. 7 , the detection position of the active block Bx closest to the car door 13 changes from yn to yn−1, indicating that the user is approaching the car door 13 .

Next, the operation of this system will be described in detail.

FIG8 is a flowchart showing the overall processing flow in this system.

When the passenger car 11 arrives at the elevator lobby 15 of any floor (Yes in step S11), the car control device 30 opens the car door 13 and waits for a user to board the passenger car 11 (step S12).

At this time, a camera 12 installed above the entrance of the elevator car 11 captures images of a predetermined range (L1) on the elevator lobby side and a predetermined range (L2) inside the car at a predetermined frame rate (e.g., 30 frames per second). The image processing device 20 acquires the images captured by the camera 12 in time series, sequentially storing them in the storage unit 21 (step S13), while executing the following user detection processing in real time (step S14).

The user detection process is executed by the user detection unit 22 provided in the image processing device 20. The user detection process is divided into a movement detection process (step S14a), a position estimation process (step S14b), and a riding intention estimation process (step S14c).

(a) Activity Detection Processing

9 is a flowchart showing the motion detection process of step S14 a. This motion detection process is executed by the motion detection unit 22 a which is one of the components of the user detection unit 22.

The motion detection unit 22a reads each image stored in the storage unit 21 one by one and calculates the average brightness value for each block (step A11). At this time, the motion detection unit 22a stores the average brightness value for each block calculated when the initial image is input in a first buffer (not shown) within the storage unit 21 as an initial value (step A12).

When the second and subsequent images are obtained, the motion detection unit 22a compares the average brightness value of each block in the current image with the average brightness value of each block in the previous image stored in the first buffer (step A13). If a block in the current image has a brightness difference greater than a predetermined value, the motion detection unit 22a determines that the block is a block with motion (step A14).

After determining whether there is motion in the current image, the motion detector 22 a stores the average brightness value of each block of the image in the first buffer for comparison with the next image (step A15 ).

Thereafter, the motion detection unit 22 a repeats the operation of determining the presence or absence of motion while sequentially comparing the brightness values of the images captured by the camera 12 in units of blocks in a time series.

(b) Position estimation processing

10 is a flowchart showing the position estimation process of step S14b. This position estimation process is executed by the position estimation unit 22b which is one of the components of the user detection unit 22.

The position estimation unit 22b checks for blocks in the current image that have motion based on the detection results of the motion detection unit 22a (step B11). If a block with motion exists within the position estimation area E1 shown in FIG4, the user detection unit 22 extracts the block closest to the car door 13 from among the blocks with motion (step B12).

Here, as shown in FIG1 , the camera 12 is installed above the entrance and exit of the elevator car 11, facing the elevator lobby 15. Therefore, when a user walks from the elevator lobby 15 toward the car door 13, there is a high probability that a portion of the user's right or left foot will be reflected in the block immediately in front of the captured image, i.e., on the side of the car door 13. Therefore, the position estimating unit 22b calculates the Y coordinate of the block closest to the car door 13 (the coordinate from the center of the car door 13 to the elevator lobby 15) as data on the position of the user's foot and stores it in a second buffer (not shown) within the storage unit 21 (step B13).

Thereafter, the position estimating unit 22b similarly calculates the Y coordinate of the block with movement closest to the car door 13 for each of the images as data on the user's foot position, and stores the data in a second buffer (not shown) within the storage unit 21. This foot position estimation process is performed not only within the position estimation area E1 but also within the boarding intention estimation area E2.

(c) Inference of boarding intention

11 is a flowchart showing the riding intention estimation process of step S14 c . This riding intention estimation process is executed by the riding intention estimation unit 22 c , which is one of the components of the user detection unit 22 .

The riding intention estimation unit 22c smoothes the data of the user's foot position in each image stored in the second buffer (step C11). As a smoothing method, a commonly known method such as a mean filter method or a Kalman filter method is used, and its detailed description is omitted here.

When smoothing the foot position data, if there is data with a change greater than a predetermined value (YES in step C12), the riding intention estimation unit 22c removes the data as an outlier (step C13). The predetermined value is determined by the user's standard walking speed and the frame rate of the captured image. Alternatively, outliers may be identified and removed before smoothing the foot position data.

Figure 12 shows the changing state of the foot position. The horizontal axis represents time, and the vertical axis represents position (Y coordinate value). When a user walks from the elevator lobby 15 toward the car door 13, the Y coordinate value of the user's foot position will gradually decrease as time passes.

Furthermore, for example, in the case of a mobile object such as a wheelchair, the data changes linearly, as shown by the dotted line. In the case of a user, the left and right feet are detected alternately, resulting in a curved data change, as shown by the solid line. Furthermore, when noise is present in the detection results, the instantaneous change in foot position increases. Data with such large changes in foot position are discarded as outliers.

Here, the boarding intention estimating unit 22c checks the change in the foot position (data change) within the boarding intention estimation area E2 shown in FIG12 (step C14). As a result, if the change in the foot position (data change) of the user walking toward the car door 13 along the Y-axis direction can be confirmed within the boarding intention estimation area E2 (Yes in step C15), the boarding intention estimating unit 22c determines that the user has the intention to board (step C16).

On the other hand, if no change in the foot position of the user walking toward the car door 13 along the Y-axis is detected within the boarding intention estimation area E2 (No in step C15), the boarding intention estimation unit 22c determines that the user has no intention to board (step C17). For example, if the person crosses the front of the passenger car 11 along the X-axis, no temporal change in the foot position along the Y-axis is detected within the boarding intention estimation area E2, as shown in FIG13 . In this case, the user is determined to have no intention to board.

In this way, by considering the active block closest to the car door 13 as the user's foot position and tracking the time change of the foot position in the Y-axis direction, it is possible to estimate whether the user intends to board the car.

Returning to FIG8 , when a user intending to board is detected (Yes in step S15 ), a user detection signal is output from the image processing device 20 to the car control device 30. Upon receiving the user detection signal, the car control device 30 prohibits the closing of the car door 13 and maintains the door open (step S16 ).

Specifically, when the car door 13 reaches the fully open state, the car control device 30 begins timing the door opening time and closes the door when the time reaches a predetermined time, T minutes (e.g., 1 minute). If a user intending to board is detected during this time and a user detection signal is received, the car control device 30 stops timing and resets the timer value. Thus, the car door 13 remains open for the time T.

Furthermore, when a new user intending to board is detected during this period, the timer value is reset to zero again, thereby maintaining the open state of the car door 13 during the time T. However, if a user approaches multiple times during the time T, the car door 13 will continue to be unable to close forever. Therefore, it is preferable to set an allowable time Tx (for example, 3 minutes) and forcibly close the car door 13 after the allowable time Tx has expired.

When the counting operation of the time T minutes ends (step S17), the car control device 30 closes the car door 13 and causes the passenger car 11 to depart toward the destination floor (step S18).

Thus, according to this embodiment, by analyzing the image obtained by photographing the elevator lobby 15 using the camera 12 installed above the entrance and exit of the passenger car 11, a user walking toward the car door 13 from a place slightly away from the passenger car 11 can be detected and reflected in the door opening and closing action.

In particular, by focusing on the position of the user's feet in the captured image and tracking the temporal changes in the foot position in the direction from the car door 13 to the elevator lobby 15 (Y-axis direction), it is possible to prevent, for example, the false detection of people who are simply passing by the passenger car, thereby accurately detecting only users who intend to board the car and reflecting this in the door opening and closing operation. In this case, since the door remains open while the user intends to board the car is detected, it is possible to avoid situations where the user is about to enter the car 11 and the door starts closing, causing the user to hit the door.

(Other embodiments)

In the above embodiment, the description focuses only on the user who intends to board the car 11. However, when the car 11 arrives at the elevator lobby 15 and opens its door, there may be users exiting the car 11. When a user exits the car 11, the position of the user's feet is detected in the direction from the car door 13 to the elevator lobby 15 (Y-axis direction), so there is a possibility that the user will be mistakenly detected as a user who intends to board the car 11.

Generally speaking, when there are users exiting the elevator car 11 and users intending to board the elevator car 11, the two users often separately board and exit the elevator on the left and right sides of the elevator car door 13. Therefore, as shown in FIG14 , a configuration is provided in which regions E3 and E4 are pre-specified, which are divided into two regions in the X-axis direction with the center of the elevator car door 13 as a reference. When the user detection unit 22 shown in FIG1 detects a block with movement in the captured image, the detection operation is performed with attention to the region with movement of a person or object toward the elevator car door 13.

Using the example of Figure 14 as an example, motion is detected in the image portion of a user attempting to board the elevator car 11 within region E3 of the captured image, and in the image portion of a user exiting the elevator car 11 within region E4. In this case, the block closest to the elevator door 13 is block B4 within image portion P4, while the block directly facing the elevator door 13 is block B3 within image portion P3. Therefore, by ignoring block B4 and focusing on region E3 and tracking the position of block B3, the user attempting to board the elevator car is detected.

In addition, in the example of Figure 14, the captured image is divided into about two areas, but the captured image can also be divided into multiple areas, focusing on areas with people/objects moving toward the elevator door 13 in these areas to detect users who intend to ride.

In this way, if we focus on the area where people/objects move toward the car door 13, we can accurately detect users who want to board the passenger car 11 and reflect this in the door opening and closing action without mistakenly detecting users coming out of the passenger car 11.

In addition, when the camera 12 is installed in the car 11, it may be necessary to install it at an angle depending on the state of the installation location. In such a case, there will be a deviation between the coordinate system of the real space shown in Figure 5 and the coordinate system on the captured image, making it impossible to accurately detect the user's movement.

Therefore, when the camera 12 is tilted, as shown in FIG15 , the tilt of each image is corrected with respect to the Y-axis in real space based on the camera 12's installation angle (tilt angle) α, and then motion detection is performed. Alternatively, as shown in FIG16 , the tilt of each image on the Y-axis is corrected based on the camera 12's installation angle (tilt angle) α, and then motion detection is performed.

Furthermore, in the above embodiment, the car door 13 of the elevator car 11 is assumed to be open in the elevator lobby 15. However, even when the car door 13 is closing, the presence of a user intending to board the car is detected using the image captured by the camera 12. When a user intending to board the car is detected, the door opening and closing control unit 31 of the car control device 30 interrupts the closing operation of the car door 13 and resumes the door opening operation.

Next, the processing actions during the door closing process will be described with reference to the flowchart of FIG17 .

When a predetermined time has passed since the car door 13 of the passenger car 11 was fully opened, the door opening and closing control unit 31 starts closing the door (step S21). At this time, the camera 12 continues to capture images. The image processing device 20 acquires images captured by the camera 12 in a time series, sequentially stores these images in the storage unit 21 (step S22), and performs user detection processing in real time (step S23).

The user detection process is performed by the user detection unit 22 included in the image processing device 20. This user detection process is divided into a motion detection process (step S23a), a position estimation process (step S23b), and a boarding intention estimation process (step S23c). These processes are similar to steps S14a, S14b, and S14c in Figure 8, so their detailed description is omitted.

Here, when a user intending to board the car is detected (Yes in step S24), a user detection signal is output from the image processing device 20 to the car control device 30. When the user detection signal is received during the door closing process, the car control device 30 interrupts the closing operation of the car door 13 and performs the door opening operation again (reopening) (step S25).

After that, the process returns to step S12 in FIG8 and the same process as above is repeated. However, if a user intending to board is continuously detected during the door closing process, the door will be repeatedly reopened, delaying the departure of the elevator car 11. Therefore, it is preferable that even if a user intending to board is detected, the door is closed instead of reopened if the allowed time Tx (e.g., 3 minutes) has expired.

Like this, even also can detect whether the user of taking intention is arranged in the door closing process, and it is reflected in the door opening and closing action.Thereby, can avoid the user from bumping into the such situation of door when wanting to enter the passenger car 11 being closed.

According to at least one embodiment described above, it is possible to provide an elevator occupancy detection system capable of accurately detecting users intending to board an elevator over a wide range and reflecting the result in the door opening and closing control.

Furthermore, while several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be implemented in various other ways and may be omitted, replaced, or modified without departing from the spirit of the invention. These embodiments and their variations are intended to be within the scope and spirit of the invention and are within the scope of the invention set forth in the claims and their equivalents.

Claims (8)

1. An elevator passenger detection system, characterized in that it comprises: The camera unit is capable of taking pictures of a specified area from near the door of the passenger car toward the waiting hall when the passenger car has arrived at the waiting hall; The user detection unit uses multiple images captured sequentially in time by the camera unit to observe the activity of people/objects within a pre-defined area, in order to detect the presence or absence of users intending to ride the vehicle; and The control unit controls the opening and closing of the door based on the detection results from the user detection unit. The width of the area is at least greater than the width of the front of the door. The user detection unit has: The activity detection unit detects human/object activity by comparing the brightness of each of the images in block units; The location estimation unit extracts active blocks detected by the activity detection unit for each image and estimates the user's location based on the coordinates of that block from the center of the door to the waiting hall; and The ride intention estimation unit infers whether a user intends to ride based on the time-series changes in location estimated by the location estimation unit. The ride intention estimation unit removes outliers from the time series changes of the positions estimated by the position estimation unit, where the changes have a pre-set value or higher. 2. The elevator passenger detection system according to claim 1, characterized in that, The position estimation unit estimates the coordinates from the center of the door to the elevator lobby in the active area as the user's foot position. If the system detects that the user's feet are in the aforementioned area and approaching the door, the ride intention estimation unit determines that the user has the intention to ride. 3. The elevator passenger detection system according to claim 1, characterized in that, The user detection unit divides each image into multiple regions based on the center of the door, focusing on areas where people/objects are moving toward the door, in order to detect the presence or absence of users who intend to ride the door. 4. The elevator passenger detection system according to claim 3, characterized in that, The user detection unit divides each image into left and right regions based on the center of the door. 5. The elevator passenger detection system according to claim 1, characterized in that, The camera unit is located above the entrance/exit of the passenger car and captures images with the direction horizontal to the door as the X-axis, the direction from the center of the door to the waiting hall as the Y-axis, and the height direction of the passenger car as the Z-axis. 6. The elevator passenger detection system according to claim 5, characterized in that, The user detection unit, after correcting the tilt of each image based on the Y-axis according to the setting angle of the camera unit, focuses on the activity of people/objects facing the door to detect whether a user intends to ride the door. 7. The elevator passenger detection system according to claim 5, characterized in that, The user detection unit, after correcting the tilt of the Y-axis on each image according to the setting angle of the camera unit, focuses on the activity of people/objects facing the door to detect whether a user intends to ride the door. 8. An elevator passenger detection system, characterized in that it comprises: The camera unit is capable of taking pictures of a specified area from near the door of the passenger car toward the waiting hall when the passenger car has arrived at the waiting hall; The user detection unit uses multiple images captured sequentially in time by the camera unit to observe the activity of people/objects within a pre-defined area, in order to detect the presence or absence of users intending to ride the vehicle; and The control unit controls the opening and closing of the door based on the detection results from the user detection unit. The user detection unit has: The activity detection unit detects active blocks by comparing the brightness of each of the images in block units. The position estimation unit extracts active regions detected by the activity detection unit for each of the images and estimates the coordinates of the active regions as the user's foot position; and The ride intention estimation unit infers whether a user intends to ride based on the time-series changes in location estimated by the location estimation unit. The ride intention estimation unit removes outliers from the time series changes of the positions estimated by the position estimation unit, where the changes have a pre-set value or higher.