JP6828108B1 - Elevator user detection system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect a user or an object by accurately setting a detection area in the vicinity of a door pocket of a door even if an adjustment work is not required and a camera mounting position is displaced. An elevator user detection system according to an embodiment includes an image pickup means, a setting target detection means, a detection area setting means, and a detection processing means. The imaging means captures a predetermined range including the vicinity of the doorway where the door opens and closes from inside the car. The setting target detecting means detects an area in which front pillars provided on at least one of both sides of the entrance / exit are reflected on the captured image obtained from the imaging means as a setting target of the detection area. The detection area setting means sets the detection area in the area where the front pillar detected by the setting target detection means is reflected. The detection processing means detects the presence or absence of a user or an object based on the image in the detection area set by the detection area setting means. [Selection diagram] Fig. 1

Description

An embodiment of the present invention relates to an elevator user detection system.

When the elevator car opens, the user's fingers may be pulled into the door pocket of the door. In order to prevent such an accident, for example, a system in which a camera is installed on the upper part of the car and a user near the door is detected and warned from the image taken by the camera is considered.

Japanese Unexamined Patent Publication No. 10-152277 Japanese Unexamined Patent Publication No. 2007-314284 Japanese Patent No. 6377795

However, when a camera is used to detect a user near the door, it is necessary to set a detection area near the door on the image taken by the camera. Usually, a method of setting the detection area based on the position of the marker placed near the door or the sill of the door is used. However, it is difficult to set the detection area accurately even with reference to the position of the marker or sill, and adjustment is required many times. In addition, if the camera mounting position shifts, the detection area will move out of the vicinity of the door, making it impossible to detect the user correctly.

The problem to be solved by the present invention is that no adjustment work is required, and even if the mounting position of the camera is displaced, the detection area is accurately set near the door pocket of the door to detect the user or an object. It is to provide a user detection system for elevators that can be used.

The elevator user detection system according to the embodiment includes an image pickup means, a setting target detection means, a detection area setting means, and a detection processing means.

The imaging means captures a predetermined range including the vicinity of the doorway where the door opens and closes from inside the car. The setting target detecting means detects a region in which a front pillar provided on at least one of both sides of the entrance / exit is reflected on a captured image obtained from the imaging means as a setting target of a detection area. The detection area setting means sets the detection area in the area where the front pillar detected by the setting target detection means is reflected. The detection processing means detects the presence or absence of a user or an object based on the image in the detection area set by the detection area setting means.
In the above configuration, the setting target detecting means displays the front pillar on the captured image based on the relative position of the imaging means with respect to the front pillar, the angle and angle of view of the imaging means, and the center coordinates on the image. A processing area having a predetermined margin is set in the area, the front pillar is specified based on the edge information of the image in the processing area, and the image in the processing area is radially radial from the center coordinates of the captured image. When there are two or more straight lines extending radially from the center coordinates of the captured image, the straight line extending to is detected as an edge indicating both ends of the inner side surface of the front pillar, and the inner side surface of the front pillar given in advance. It is characterized in that two straight lines close to the center line of the above are selected as edges indicating both ends of the inner side surface of the front pillar.

FIG. 1 is a diagram showing a configuration of an elevator user detection system according to an embodiment. FIG. 2 is a diagram showing a configuration of a portion around an entrance / exit in the car according to the same embodiment. FIG. 3 is a diagram showing an example of a photographed image of the camera in the same embodiment. FIG. 4 is a flowchart showing the overall processing flow of the user detection system in the same embodiment. FIG. 5 is a diagram for explaining a coordinate system in real space in the same embodiment. FIG. 6 is a diagram for explaining the relationship between the center coordinates on the image and the structure reflected in the image in the same embodiment. FIG. 7 is a flowchart showing details of the setting target detection process executed in step S10 of FIG. FIG. 8 is a diagram showing an example of a captured image used for the setting target detection process in the same embodiment. FIG. 9 is a diagram schematically showing an edge detection result of the captured image of FIG. FIG. 10 is a diagram for explaining a method for detecting a region in which the front pillar is reflected in the car from the captured image of FIG. FIG. 11 is a diagram showing a state in which the area in which the front pillar is reflected in the car is specified. FIG. 12 is a diagram for explaining a method of detecting an area in which the front pillar is reflected by using an image taken when the car door is closed in the same embodiment. FIG. 13 is a diagram showing the relationship between the user in the car and the detection area in the same embodiment. FIG. 14 is a diagram showing the relationship between the user and the detection area on the captured image in the same embodiment. FIG. 15 is a diagram for explaining the difference method used for the user detection process in the same embodiment. FIG. 16 is a diagram for explaining motion detection used in the user detection process in the same embodiment. FIG. 17 is a diagram for explaining boundary detection used in the user detection process in the same embodiment. FIG. 18 is a diagram showing a configuration of a portion around an entrance / exit in a car in which a single-door type car door is used in the same embodiment. FIG. 19 is a diagram for explaining an opening / closing operation of a single-door type car door in the same embodiment.

Hereinafter, embodiments will be described with reference to the drawings.
The disclosure is merely an example, and the invention is not limited by the contents described in the following embodiments. Modifications that can be easily conceived by those skilled in the art are naturally included in the scope of disclosure. In order to clarify the explanation, in the drawings, the size, shape, etc. of each part may be changed with respect to the actual embodiment and represented schematically. In a plurality of drawings, the corresponding elements may be given the same reference numbers and detailed description may be omitted.

FIG. 1 is a diagram showing a configuration of an elevator user detection system according to an embodiment. Although the description will be given using one car as an example, the same configuration is used for a plurality of cars.

A camera 12 is installed above the doorway of the car 11. Specifically, the camera 12 is installed in the curtain plate 11a covering the upper part of the entrance / exit of the car 11 with the lens portion tilted directly downward, or on the landing 15 side or the inside of the car 11 by a predetermined angle. ..

The camera 12 is a small surveillance camera such as an in-vehicle camera, has a wide-angle lens or a fisheye lens, and can continuously capture images of several frames (for example, 30 frames / second) per second. The camera 12 is activated when the car 11 arrives at the landing 15 on each floor, and photographs a predetermined range L including the vicinity of the car door 13.

The location of the camera 12 does not have to be the upper part of the doorway of the car 11 as long as it is near the car door 13. For example, it may be a place where the vicinity of the entrance / exit of the car 11 can be photographed, such as the upper part of the side wall near the entrance / exit of the car 11. By installing the camera 12 in such a place, the detection area described later can be set appropriately, and the user or an object can be correctly detected from the image in the detection area.

On the other hand, since surveillance cameras generally used for surveillance purposes are installed in the back of the car or on the ceiling surface, the shooting range is wide throughout the car. For this reason, it is difficult to set the detection area, and there is a high possibility that even a user who is far from the entrance / exit of the car 11 will be detected.

At the landing 15 on each floor, a landing door 14 is openably and closably installed at the arrival port of the car 11. The landing door 14 engages with the car door 13 when the car 11 arrives to open and close. The power source (door motor) is on the car 11 side, and the landing door 14 only opens and closes following the car door 13. In the following description, it is assumed that the landing door 14 is also open when the car door 13 is open, and the landing door 14 is also closed when the car door 13 is closed.

Each image (video) continuously captured by the camera 12 is analyzed and processed in real time by the image processing device 20. Although the image processing device 20 is taken out from the car 11 for convenience in FIG. 1, the image processing device 20 is actually housed in the curtain plate 11a together with the camera 12.

The image processing device 20 is provided with a storage unit 21 and a detection unit 22. The storage unit 21 has a buffer area for sequentially storing images taken by the camera 12 and temporarily storing data necessary for processing of the detection unit 22. It should be noted that the storage unit 21 may store an image that has undergone processing such as distortion correction, enlargement / reduction, and partial cropping as preprocessing for the captured image.

The detection unit 22 detects a user near the car door 13 by using the captured image of the camera 12. When the detection unit 22 is functionally divided, it is composed of a setting target detection unit 22a, a detection area setting unit 22b, and a detection processing unit 22c.

The setting target detection unit 22a detects an area on the captured image obtained from the camera 12 in which the front pillars provided on at least one of the entrances and exits of the car 11 are reflected as the setting target of the detection area.

The detection area setting unit 22b sets the detection area in the area where the front pillar detected by the setting target detection unit 22a is reflected. Specifically, the detection area setting unit 22b sets a band-shaped detection area along the inner side surface of the front pillar on the captured image. The "front pillar" is also referred to as an entrance / exit pillar or an entrance / exit frame, and is provided on both sides or one side of the entrance / exit of the car 11 (see FIG. 2). A door pocket for storing the car door 13 is generally provided on the back side of the front pillar.

The detection processing unit 22c detects the presence or absence of a user or an object based on the image in the detection area set by the detection area setting unit 22b. The "thing" referred to here includes, for example, a user's clothes and luggage, and a moving body such as a wheelchair. The car control device 30 may have some or all of the functions of the image processing device 20.

The car control device 30 controls the operation of various devices (destination floor buttons, lighting, etc.) installed in the car 11. Further, the car control device 30 includes a door opening / closing control unit 31 and a notification unit 32. The door opening / closing control unit 31 controls the opening / closing of the car door 13 when the car 11 arrives at the landing 15. Specifically, the door opening / closing control unit 31 opens the car door 13 when the car 11 arrives at the landing 15, and closes the door after a predetermined time has elapsed.

Here, when a user or an object is detected by the detection processing unit 22c while the door of the car door 13 is open, the door opening / closing control unit 31 is for avoiding a door accident (accident of being pulled into the door pocket). Door open / close control is performed. Specifically, the door opening / closing control unit 31 suspends the door opening operation of the car door 13, moves it in the opposite direction (door closing direction), or slows down the door opening speed of the car door 13. The notification unit 32 calls attention to the user in the car 11 based on the detection result by the detection processing unit 22c.

FIG. 2 is a diagram showing a configuration of a portion around the entrance / exit in the car 11.

A car door 13 is provided at the entrance and exit of the car 11 so as to be openable and closable. In the example of FIG. 2, a double door double door type car door 13 is shown, and the two door panels 13a and 13b constituting the car door 13 open and close in opposite directions along the frontage direction (horizontal direction). .. The "frontage" is the same as the entrance / exit of the car 11.

Front pillars 41a and 41b are provided on both sides of the doorway of the car 11, and surround the doorway of the car 11 together with the curtain plate 11a. When the car door 13 is opened, one door panel 13a is housed in a door pocket 42a provided on the back side of the front pillar 41a, and the other door panel 13b is housed in the door pocket 42b provided on the back side of the front pillar 41b. ..

A display 43, an operation panel 45 on which a destination floor button 44 and the like are arranged, and a speaker 46 are installed on one or both of the front pillars 41a and 41b. In the example of FIG. 2, the speaker 46 is installed on the front pillar 41a, the display 43 and the operation panel 45 are installed on the front pillar 41b.

Here, the camera 12 is installed in the central portion of the curtain plate 11a above the doorway of the car 11. The camera 12 is installed from the lower part of the curtain plate 11a downward so that when the car door 13 is opened together with the landing door 14, the photograph including the vicinity of the doorway can be taken (see FIG. 3).

FIG. 3 is a diagram showing an example of a captured image of the camera 12. It shows a case where the car door 13 (door panels 13a and 13b) and the landing door 14 (door panels 14a and 14b) are fully opened and the image is taken downward from the upper part of the doorway of the car 11. In FIG. 3, the upper side shows the landing 15 and the lower side shows the inside of the car 11.

At the landing 15, three-sided frames 17a and 17b are provided on both sides of the arrival port of the car 11, and a band-shaped landing sill 18 having a predetermined width is provided on the floor surface 16 between the three-sided frames 17a and 17b. It is arranged along the opening / closing direction of 14. Further, a strip-shaped car sill 47 having a predetermined width is arranged on the entrance / exit side of the floor surface 16 of the car 11 along the opening / closing direction of the car door 13.

Here, the detection areas Ea and Eb are set on the inner side surfaces 41a-1 and 41b-1 of the front pillars 41a and 41b on the captured image, respectively. The detection areas Ea and Eb are areas for detecting a user or an object on the captured image, and are used here to prevent an accident of being pulled into the door pockets 42a and 42b during the door opening operation.

The detection areas Ea and Eb have predetermined widths D1 and D2 in the width direction of the inner side surfaces 41a-1 and 41b-1 of the front pillars 41a and 41b, respectively, and are set in a band shape. The widths D1 and D2 are set to be the same as or slightly smaller than the width (width in the lateral direction) of the inner side surfaces 41a-1 and 41b-1, for example. The values may be the same as or different from the widths D1 and D2.

Further, the widths D1 and D2 may be partially changed so that, for example, the widths D1a and D2a of the portion easily touched by the user may be slightly wider than the widths D1 and D2 (see FIG. 14). As a result, it is possible to detect an accident in which the door pocket is pulled in at an early stage.

The front of the front pillars 41a and 41b is outside the area setting. This is because there is an operation panel 45 or the like in front of the front pillars 41a and 41b, and users are often nearby. If the inner side surfaces 41a-1 and 41b-1 of the front pillars 41a and 41b are used, the user who operates the operation panel 45 is not erroneously detected, and the detection area Ea is not affected by the opening / closing operation of the car door 13. , Eb can be set.

Next, the operation of this system will be described in detail.
FIG. 4 is a flowchart showing the overall processing flow in this system.

First, as initial settings, the setting target detection unit 22a of the detection unit 22 provided in the image processing device 20 executes the setting target detection process, and the detection area setting unit 22b executes the detection area setting process (steps S10-S11). .. These processes are executed as follows, for example, when the camera 12 is installed or when the installation position of the camera 12 is adjusted.

(Setting target detection process)
The setting target detection process is a process of detecting a setting target of a detection area designated in advance on a captured image. In the present embodiment, the front columns 41a and 41b are targeted for setting the detection area. The area where the front pillars 41a and 41b are projected on the captured image is calculated based on the following camera information.

-Relative position of the camera with respect to the front pillar (3D coordinates)
・ Camera angle (3 axes)
・ Camera angle of view (focal length)
・ Center coordinates on the image

As shown in FIG. 5, "three-dimensional coordinates" means that the direction horizontal to the car door 13 is the X-axis, and the direction from the center of the car door 13 to the landing 15 (the direction perpendicular to the car door 13) is Y. These are the coordinates when the axis is the axis and the height direction of the car 11 is the Z axis. Further, the "center coordinates on the image" are two-dimensional coordinate positions where the lens optical axis of the camera 12 passes over the captured image.

Here, as shown in FIG. 6, when the center coordinates on the image are Po (xo, yo), the Z direction (vertical direction) of the elevator structures 51 and 52 reflected in the photographed image in the three-dimensional space. ) Is reflected as a straight line extending radially from the center coordinate Po. The setting target detection unit 22a detects the area where the front pillars 41a and 41b are reflected from the captured image by utilizing such characteristics of the image.

The setting target detection process executed in step S10 will be described in detail below.
FIG. 7 is a flowchart showing the processing operation of the setting target detection unit 22a.

Now, as shown in FIG. 8, it is assumed that the image taken by the camera 12 is used with the car door 13 and the landing door 14 opened. The upper side shows the landing 15 and the lower side shows the inside of the car 11. Reference numerals 53 and 54 in the figure are arbitrary objects (such as luggage) placed on the floor surface 16 of the landing 15. In practice, such objects 53, 54 are not placed near the doorway of the car 11, but here for comparison with elevator structures such as the three-sided frames 17a, 17b and the front columns 41a, 41b. It is shown for convenience.

In the following, the case where the detection area Ea is set on one front pillar 41a in the car 11 will be described, but the same applies to the case where the detection area Eb is set on the other front pillar 41b.

The setting target detection unit 22a is given the above-mentioned camera information, that is, information such as the camera relative position with respect to the front pillar, the angle of the camera, the angle of view, and the center coordinates. Based on this information, the setting target detection unit 22a sets a processing area 61 in which the front pillar 41a is projected on the captured image with predetermined margins in the width direction and the height direction of the front pillar 41a. (Step S21).

The processing area 61 is called "ROI (Region Of Interest)", and indicates an area in which the detection area setting target (front pillars 41a, 41b) is likely to be reflected on the captured image. The reason why the processing area 61 has a predetermined margin is to consider the possibility that an error may occur between the camera information given in advance and the actual camera information, for example, because the mounting position of the camera 12 shifts during operation. That's it.

When such a processing area 61 is set on the captured image, the setting target detection unit 22a detects an edge from the image in the processing area 61 (step S22). “Edge” means a boundary line between regions having different characteristics such as color, brightness, and pattern, in addition to straight lines and curves in the image. This edge detection may be performed by a known method. For example, a well-known image processing technique such as a Laplacian filter or the Canny method may be used, or the boundary position of the region may be obtained by the difference in the dispersion of the brightness values of the pixels.

FIG. 9 schematically shows the edge detection result of the captured image of FIG. The part indicated by the white line is the edge. In the example of FIG. 9, the straight line of the edge is clearly expressed, but in reality, the straight line of the edge is generally missing in some places due to the influence of the pattern and scratches on the floor and wall. is there. Noise such as patterns and scratches can be eliminated based on the edge strength information generally obtained by the above edge detection method. Further, in the example of FIG. 9, although the edge of the entire captured image is shown, only the edge of the image in the processing area 61 may be detected.

The front pillar 41a in the car 11 is erected vertically from the floor surface 19 toward the ceiling in the three-dimensional space. In the two-dimensional image, both ends of the inner side surface 41a-1 of the front pillar 41a are reflected as straight lines extending radially from the center coordinate Po of the image. Therefore, the setting target detection unit 22a is set at both ends of the inner side surface 41a-1 of the front pillar 41a based on the edge information of the image in the processing area 61 under the condition of "a straight line extending radially from the center coordinate Po". The edge (two straight lines) indicating the above is detected (step S23).

Here, even if the condition of "a straight line extending radially from the center coordinate Po" is satisfied, edges that are not both ends of the inner side surface 41a-1 of the front pillar 41a may be detected in the image in the processing area 61. is there. In particular, when the car 11 is opened, the tip of the door panel 13a is included in the processing region 61 as shown in FIG. 8, so that the edge of that portion is detected as an edge satisfying the above conditions. Further, for example, when the inner side surface 41a-1 of the front pillar 41a has a linear pattern or a scratch, it is detected as an edge satisfying the above conditions.

When two or more straight lines (edges) satisfying the above conditions are detected, as shown in FIG. 10, in the edge information detected from the image in the processing area 61, the inner side surface 41a- of the front pillar 41a With reference to the center line 62 of 1, two straight lines 63 and 64 close to the center line 62 may be selected. The straight lines 63 and 64 are lines extending in the Z direction (vertical direction) from the floor surface 19 in a three-dimensional real space. The position coordinates of the center line 62 are obtained based on the above-mentioned camera information.

Returning to FIG. 7, when two or more straight lines (edges) satisfying the above conditions are detected from the image in the processing area 61 (Yes in step S24), the setting target detection unit 22a is centered in advance. Based on the position coordinates of the line 62, the two straight lines 63 and 64 near the center line 62 are selected as the edges indicating the inner side surface 41a-1 of the front pillar 41a (step S25).

When the two straight lines 63 and 64 are obtained in this way, the setting target detection unit 22a has the inside of the rectangle connecting the upper and lower sides of the straight lines 63 and 64 (the shaded portion 65 in FIG. 11) inside the front pillar 41a. It is specified as an area where the side surface 41a-1 is reflected (step S26).

The same applies to the inner side surface 41b-1 of the front pillar 41b, and the inner side surface 41b of the front pillar 41b is based on the edge information of the image in the processing area (ROI) set near the inner side surface 41b-1 of the front pillar 41b. The area where -1 is reflected is specified.

(Detection area setting process)
The detection area setting process is a process of setting a detection area for detecting a user or an object on a captured image. The detection area setting unit 22b receives the result detected by the setting target detection unit 22a, and sets the detection areas Ea and Eb in the area where the front pillars 41a and 41b are reflected on the captured image (step S11). Specifically, the detection area setting unit 22b sets the detection area Ea in the area where the inner side surface 41a-1 of the front pillar 41a is reflected on the captured image. Similarly, the detection area setting unit 22b sets the detection area Eb in the area where the inner side surface 41b-1 of the front pillar 41b is reflected on the captured image.

In this case, as shown in FIG. 11, the entire region (hatched portion) detected by the setting target detection unit 22a may be the detection area Ea, or a range of a certain height may be the detection area Ea. The relationship between the height on the image and the height in the real space is calculated using the above-mentioned camera information. Further, the width of the detection area Ea may also be provided with a margin in the direction of reducing by a certain amount. Further, the width of the detection area Ea may be partially changed (see FIG. 14). The same applies to the detection area Eb.

By setting the detection areas Ea and Eb on the inner side surfaces 41a-1 and 41b-1 of the front pillars 41a and 41b in this way, for example, as shown in FIG. 13, the user can use the front pillars 41a when the door is opened. When a hand is touched on the inner side surface 41a-1, it is possible to detect the hand before it is pulled into the door pocket 42a.

(Detection of front pillars using images taken when the door is closed)
In the examples of FIGS. 8 to 11, the case where the front pillar 41a is detected by using the image taken when the door of the car 11 is opened has been described, but the image taken when the door of the car 11 is closed can also be used. good.

FIG. 12 shows an example of an image taken when the car 11 is closed.
Even when an image taken when the door of the car 11 is closed is used, a processing area in which the front pillar 41a is reflected on the shot image has predetermined margins in the width direction and the height direction of the front pillar 41a ( ROI) 71 may be set, and the inner side surface 41a-1 of the front pillar 41a may be detected from the edge information of the image in the processing area 71.

In this case, since the tip of the door panel 13a is not reflected in the image in the processing area 71, the two straight lines 73 and 74 close to the center line 72 of the inner side surface 41a-1 are formed on the inner side surface 41a-1 of the front pillar 41a. It is easy to identify as an edge indicating both ends of. The same applies to the case of detecting the inner side surface 41b-1 of the other front pillar 41b.

The operation of the car 11 during operation will be described below.
As shown in FIG. 4, when the car 11 arrives at the landing 15 on an arbitrary floor (Yes in step S12), the car control device 30 opens the car door 13 (step S13).

At this time (during the door opening operation of the car door 13), the periphery of the car door 13 (front pillars 41a, 41b, etc.) is set to a predetermined frame rate (for example, 30 frames / second) by the camera 12 installed at the upper part of the doorway of the car 11. ) Is taken. The image processing device 20 acquires images taken by the camera 12 in time series, and while sequentially storing these images in the storage unit 21 (step S14), executes the following user detection processing in real time. (Step S15). As preprocessing for the captured image, distortion correction, enlargement / reduction, and partial cropping of the image may be performed.

The user detection process is executed by the detection processing unit 22c of the detection unit 22 provided in the image processing device 20.

That is, the detection processing unit 22c detects the presence or absence of a user or an object based on these images by extracting the images in the detection areas Ea and Eb from the plurality of captured images obtained by the camera 12 in time series. .. Specifically, it is detected by the following method.

(A) Difference Method As shown in FIG. 15, the detection processing unit 22c compares the images in the detection areas Ea and Eb with the basic images in chronological order, and determines the presence or absence of a user or an object from the difference between the two images. Detect. FIG. 15A is a basic image, which is obtained by extracting the images in the detection areas Ea and Eb from the images taken by the camera 12 in the state where there are no users or objects in the car 11 in advance. FIG. 3B is a detection target image, which is obtained by extracting images in the detection areas Ea and Eb from the captured image when the door is opened.

The detection processing unit 22c compares the basic image and the detection target image, and if the two images have a predetermined amount or more of different pixel values, a user or an object exists near the door pockets 42a and 42b. Judge that.

(B) Motion detection As shown in FIG. 16, the detection processing unit 22c divides the captured image into a matrix in predetermined block units, and pays attention to the moving blocks among these blocks, and the user or the object. Detects the presence or absence of.

Specifically, the detection processing unit 22c reads out each image stored in the storage unit 21 one by one in chronological order, and calculates the average luminance value of these images for each block. At that time, the average luminance value for each block calculated when the first image is input as the initial value is held in the first buffer area (not shown) in the storage unit 21.

When the second and subsequent images are obtained, the detection processing unit 22c determines the average luminance value for each block of the current image and the average luminance value for each block of the previous image held in the first buffer area. Compare with. As a result, when there is a block having a brightness difference equal to or larger than a preset value in the current image, the detection processing unit 22c determines the block as a moving block. Upon determining the presence or absence of motion with respect to the current image, the detection processing unit 22c holds the average luminance value for each block of the image in the first buffer area for comparison with the next image. After that, in the same manner, the detection processing unit 22c repeats determining the presence or absence of movement while comparing the brightness values of each image in block units in chronological order.

The detection processing unit 22c checks whether or not there is a moving block in the images in the detection areas Ea and Eb. As a result, if there is a moving block in the image in the detection areas Ea and Eb, the detection processing unit 22c determines that a person or an object exists near the door pockets 42a and 42b.

As shown in FIG. 3, the detection areas Ea and Eb are set on the inner side surfaces 41a-1 and 41b-1 of the front pillars 41a and 41b. Therefore, the movement when the car doors 13 (door panels 13a and 13b) are opened and closed is not detected in the detection areas Ea and Eb.

(C) Boundary detection The detection processing unit 22c detects the boundary of the elevator structure from the images in the detection areas Ea and Eb. The "boundary of the elevator structure" referred to here is a boundary between the inner side surfaces 41a-1, 41b-1 of the front columns 41a and 41b and the door pockets 42a and 42b. When the boundary is interrupted on the image (when it is partially hidden), the detection processing unit 22c determines that a user or an object exists.

In this method, it is necessary to expand the detection areas Ea and Eb as shown in FIG. 17 and set them to include the above boundary. Since the method of detecting the boundary in the detection area on the image is known, for example, in Japanese Patent Application Laid-Open No. 2017-240799, detailed description thereof will be omitted here.

In the image taken by the camera 12, there is a boundary between the inner side surfaces 41a-1, 41b-1 of the front pillars 41a and 41b and the door pockets 42a and 42b regardless of the open / closed state of the car door 13. Therefore, by detecting whether or not the boundary is interrupted on the image, it is possible to reliably detect a user or an object close to the door pockets 42a and 42b, and a user away from the door pockets 42a and 42b. Or, there is no false detection of an object.

As another method, a structure other than the elevator structure may be recognized from the images in the detection areas Ea and Eb, and the existence of the user or the object may be determined based on the recognition result. The method of recognizing the structure may be a generally known method. For example, there are deep learning, SVM (Support Vector Machine), and random forest.

Returning to FIG. 4, when the presence of a user or an object is detected in the detection areas Ea and Eb during the door opening operation of the car door 13 (Yes in step S16), the image processing device 20 shifts to the car control device 30. On the other hand, a user detection signal is output. When the door opening / closing control unit 31 of the car control device 30 receives the user detection signal, the door opening operation of the car door 13 is temporarily stopped, and after a few seconds, the door opening operation is restarted from the stopped position (step S17).

When the user detection signal is received, the door opening speed of the car door 13 may be slower than usual, or the car door 13 may be slightly moved in the opposite direction (door closing direction) before the door opening operation is performed. It may be restarted.

Further, the notification unit 32 of the car control device 30 makes a voice announcement through the speaker 46 in the car 11 to alert the user to move away from the door pockets 42a and 42b (step S18). The notification method is not limited to voice announcement, for example, it may be possible to display a message such as "Because it is dangerous near the door pocket, please leave immediately.", Or use voice announcement and message display together. But it's okay. Further, a warning sound may be sounded.

The above process is repeated while the presence of a user or an object is detected in the detection areas Ea and Eb. As a result, for example, when the user puts his / her hand near the door pocket 42a, it is possible to prevent the user from being drawn into the door pocket 42a.

If the presence of a user or an object is not detected in the detection areas Ea and Eb (No in step S16), the car control device 30 continues the door closing operation of the car door 13, and the car is closed after the door is closed. 11 is departed toward the destination floor (step S19).

In the above embodiment, the double door type car door has been described as an example, but the same applies to the single door type as shown in FIG.

FIG. 18 is a diagram showing a configuration of a portion around an entrance / exit in a car in which a two-door single-door type car door is used. In this example, a two-door single-door type car door 13 is installed at the doorway of the car 11 so as to be openable and closable. As shown in FIG. 19, the car door 13 has two door panels 13a and 13b, which open and close in the same direction along the frontage direction.

When the car door 13 is a single door type, the door pocket 42a is provided only on one side of the doorway. In the example of FIG. 18, a door pocket 42a is provided on the left side of the doorway, and when the door is opened, the two door panels 13a and 13b are stored in the door pocket 42a in an overlapping state.

Here, the camera 12 installed on the curtain plate 11a is moved closer to the door pocket 42a side, and the detection area Ea is set for the front pillar 41a on the door pocket 42a side. Specifically, as described with reference to FIG. 3, a band-shaped detection area Ea having a predetermined width D1 is set along the inner side surface 41a-1 of the front pillar 41a. As a result, for example, when the user's hand is near the door pocket 42a, the state can be detected from the image in the detection area Ea, and by reflecting it in the door opening / closing operation, an accident of being drawn into the door pocket 42a can be prevented. Can be prevented.

If the detection area Eb is also set for the other front pillar 41b in FIG. 18, it is possible to prevent an accident (door hit accident) in which the side end of the car door 13 hits when the door is closed. ..

As described above, according to the present embodiment, the area in which the front pillars 41a and 41b in the car 11 are reflected on the captured image can be detected by image processing, and the detection areas Ea and Eb can be accurately set there. .. Therefore, for example, it is not necessary to place a marker near the door pockets 42a and 42b and adjust the set position of the detection areas Ea and Eb with reference to the position of the marker. Further, even if the mounting position of the camera 12 is displaced, the area where the front pillars 41a and 41b are actually reflected is specified by using the captured image obtained from the camera 12, and the detection areas Ea and Eb are accurately set there. Can be set.

By setting the detection areas Ea and Eb on the front pillars 41a and 41b in this way, it is possible to correctly detect a user or an object near the door, for example, an accident in which the door is pulled into a door pocket. Elevators can be used safely by preventing accidents that occur when opening and closing. On the other hand, by limiting the places where the detection areas Ea and Eb are set to the front pillars 41a and 41b, it is possible to avoid erroneous detection of a user or an object away from the door, and prevent unnecessary door control and alerting. be able to.

In the above embodiment, the detection areas Ea and Eb are set on both sides of the entrance and exit of the car 11, but the detection areas may be set on at least one of them.

According to at least one embodiment described above, no adjustment work is required, and even if the camera mounting position is displaced, the detection area is accurately set near the door pocket of the door to detect the user or an object. It is possible to provide an elevator user detection system that can be used.

In the above embodiment, the door provided in the car of the elevator has been described, but the present invention can be applied to an automatic door provided at the entrance of a building, for example. That is, for example, in the case of an automatic door at the entrance of a building, a camera is installed above the entrance, pillars on both sides of the entrance are detected on an image taken by the camera, and the detection area Ea, Set Eb (if it is one-sided, either one). After that, as in the above embodiment, the user or the object is detected by the images in the detection areas Ea and Eb and reflected in the door opening / closing control, and the attention is issued.

In short, some embodiments of the present invention have been described, but 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 embodiments, and various omissions, replacements, and changes can be made without departing from the gist of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are also included in the scope of the invention described in the claims and the equivalent scope thereof.

11 ... Riding car, 11a ... Curtain board, 12 ... Camera, 13 ... Car door, 14 ... Landing door, 15 ... Landing, 16 ... Landing floor, 17a, 17b ... Three-sided frame, 18 ... Landing sill, 19 ... Riding The floor of the car, 20 ... image processing device, 21 ... storage unit, 22 ... detection unit, 22a ... setting target detection unit, 22b ... detection area setting unit, 22c ... detection processing unit, 30 ... car control device, 31 ... door Open / close control unit, 41a, 41b ... front pillar, 41a-1, 41b-1 ... inner side surface of front pillar, 42a, 42b ... door pocket, 43 ... display, 44 ... destination floor button, 45 ... operation panel, 46 ... speaker , 47 ... Basket sill, 32 ... Notification unit, Ea, Eb ... Detection area, Po ... Center coordinates, 61 ... Processing area, 62 ... Center line, 63, 64 ... Straight line, 71 ... Processing area, 72 ... Center line, 73 , 74 ... Straight line.

Claims (8)

An imaging means that captures a predetermined range including the vicinity of the doorway where the door opens and closes from inside the car,
A setting target detecting means for detecting a region in which a front pillar provided on at least one of both sides of the entrance / exit is reflected as a setting target of a detection area on a captured image obtained from the imaging means.
The detection area setting means for setting the detection area in the area where the front pillar detected by the setting target detection means is reflected, and the detection area setting means.
It is provided with a detection processing means for detecting the presence or absence of a user or an object based on an image in the detection area set by the detection area setting means .
The above setting target detection means
A processing area in which a region in which the front pillar is projected on the captured image has a predetermined margin based on the relative position of the imaging means with respect to the front pillar, the angle of the imaging means, the angle of view, and the center coordinates on the image. Set and
Identify the front pillar based on the edge information of the image in the processing area,
A straight line extending radially from the center coordinates of the captured image on the image in the processing area is detected as an edge indicating both ends of the inner side surface of the front pillar.
When there are two or more straight lines extending radially from the center coordinates of the photographed image, two straight lines close to the center line of the inner side surface of the front pillar given in advance are applied to both ends of the inner side surface of the front pillar. An elevator user detection system characterized by being selected as the indicated edge . The above setting target detection means
The elevator user detection system according to claim 1 , wherein the front pillar detection process is performed with the door of the car closed. The above detection area setting means
The user detection system for an elevator according to claim 1, wherein the detection area is set in a part or the whole of the inner side surface of the front pillar in the height direction. The user detection system for an elevator according to claim 3 , wherein the detection area is set with a predetermined width in the width direction of the inner side surface of the front pillar. The above detection processing means
The user detection system for an elevator according to claim 1, wherein the presence or absence of a user or an object is detected based on an image in the detection area during the door opening operation of the door. The above imaging means
The elevator user detection system according to claim 1, wherein the elevator is installed above the entrance / exit of the car. The user detection system for an elevator according to claim 1, further comprising a door opening / closing control means for controlling the opening / closing operation of the door based on the detection result of the detection processing means. The user detection system for an elevator according to claim 1, further comprising a notification means for calling attention to a user in the car based on the detection result of the detection processing means.

Images