CN107055238A - Image processing apparatus - Google Patents

Abstract

Embodiments of the present invention relate to an image processing apparatus. Provided is an image processing device capable of suppressing a reduction in the detection accuracy of a hall situation caused by a deviation in the installation angle of a camera. According to an embodiment, there is provided an image processing device that performs image processing on an image including a hall captured by a camera in order to detect a situation in which a user gets in the hall of a car. An image processing apparatus according to an embodiment includes an acquisition unit, a calculation unit, and a correction unit. The acquisition unit acquires an image captured by the camera. The calculation unit calculates a difference between the position of the reference portion and a reference position determined in advance for the reference portion when the position of the reference portion included in the acquired image does not match the reference position. The correction unit corrects a setting value relating to image processing based on the calculated difference.

Description

image processing device

This application is based on Japanese Patent Application No. 2016-004829 (filing date: January 13, 2016), and priority is claimed based on this application. This application incorporates the entire contents of this application by referring to this application.

technical field

Embodiments of the present invention relate to an image processing device.

Background technique

Generally, in an elevator system, a door (hereinafter, referred to as a car door) is attached to a passenger car on which a user rides.

In such an elevator system, when a passenger car arrives (levels up) at a hall designated by a user, the car door is opened so that the user can board the passenger car.

On the other hand, when the passenger car performs a lifting operation (rising or descending), the safety of the user can be ensured by keeping the car door in the closed state.

Here, since the above-mentioned car doors are opened and closed using a motor, there is a possibility that a user may be caught by the car doors when boarding, for example.

Therefore, it is conceivable to prevent accidents in which the user is caught by the door by detecting the situation of the elevator hall (for example, a user who wants to board a car) using, for example, a camera (image captured by the camera).

However, when the installation angle of the camera (the angle of view, etc.) deviates, there is a possibility that the detection accuracy of the situation of the elevator hall will decrease.

Contents of the invention

The problem to be solved by the present invention is to provide an image processing device capable of suppressing a decrease in detection accuracy of an elevator hall situation due to a deviation in the installation angle of a camera.

According to an embodiment, there is provided an image processing device that performs image processing on an image captured by a camera including the hall in order to detect a situation in which a user boards in the hall. An image processing device according to an embodiment includes an acquisition unit, a calculation unit, and a correction unit. The acquiring unit acquires an image captured by the camera. The calculation unit calculates a difference between the position of the reference part and the reference position when the position of the reference part included in the acquired image does not coincide with a predetermined reference position for the reference part. The correction unit corrects a setting value related to the image processing based on the calculated difference.

According to the image processing device configured as described above, it is possible to suppress a decrease in the detection accuracy of the situation of the elevator hall due to the deviation of the installation angle of the camera.

Description of drawings

FIG. 1 is a diagram showing the configuration of an elevator system including an image processing device according to an embodiment.

FIG. 2 is a diagram for specifically explaining an installation angle of a camera.

FIG. 3 is a diagram showing an example of an image captured by a camera.

FIG. 4 is a block diagram showing the functional configuration of the image processing device.

FIG. 5 is a flowchart showing the processing procedure of the image processing device in the calibration function.

FIG. 6 is a diagram showing an example of an image captured by the camera in a state in which the installation angle of the camera is not shifted.

FIG. 7 is a diagram showing an example of an image captured by the camera in a state in which the camera's mounting angle is shifted.

FIG. 8 is a diagram showing an example of a corrected mask area.

detailed description

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

FIG. 1 shows the configuration of an elevator system including an image processing device according to this embodiment. In an elevator system, a passenger car 10 connected by main ropes and a counterweight (not shown) are provided in a lifting path, and are supported on guide rails (not shown) so as to be movable up and down. In addition, a hoist around which the main rope is wound is installed on the upper part of the lifting path. In an elevator system, by driving such a hoist, the passenger car 10 (and the counterweight) can be raised and lowered.

In addition, an openable and closable car door 10a is provided at the landing on the front of the passenger car 10 . Opening and closing control of the car door 10a is performed by, for example, a car control device (not shown) provided on the upper portion of the passenger car 10 .

According to such an elevator system, by opening the car door 10a when the passenger car 10 arrives (levels) at the lobby 20 designated by the user, it is possible to allow the user to board the passenger car 10. Car 10. In addition, during the raising and lowering operation of the passenger car 10, the safety of the user riding on the passenger car 10 can be ensured by making the car door 10a in the closed state.

In addition, although omitted in FIG. 1 , in order to prevent users who are waiting in the elevator hall 20 to get on the passenger car 10 from falling into the ascending and descending path, etc., the elevator hall 20 is provided with an elevator hall door. . The hall door is opened and closed in association with the opening and closing of the above-mentioned car door 10a.

In the present embodiment, a camera (imaging device) 30 that captures images including the hall 20 where the user gets on the car 10 is installed inside the car 10 . Specifically, the camera 30 is installed near the center of the side plate portion provided above the entrance of the passenger car 10, and as shown in FIG. An image of the boundary portion between the passenger car 10 and the elevator hall 20 when the elevator hall door installed in the elevator hall 20 is in an open state. In other words, the camera 30 is installed at a position capable of capturing images of the area on the elevator hall 20 side and the area on the passenger car 10 side. The camera 30 is, for example, a camera capable of capturing several frames of images per second, such as a vehicle-mounted camera.

Here, the installation angle (position) of the camera 30 will be specifically described with reference to FIG. 2 . In the present embodiment, the camera 30 is installed at a height h (for example, about 2100 mm) from the floor surface of the passenger car 10, and captures images of the range L1+L2. L1 is an imaging range on the side of the hall 20, and is a range of, for example, about 2000 to 3000 mm from the car door 10a (or the hall door). L2 is an imaging range on the side of the passenger car 10, and is, for example, a range of about 350 to 500 mm from the car door 10a to the back of the car. In addition, L1 and L2 are ranges in the depth direction, and the range in the width direction (direction perpendicular to the depth direction) is at least larger than the lateral width of the car 10 .

In the camera 30 , the angle of view (α) is set in advance so that the above-mentioned photographing range (area) can be photographed.

FIG. 3 shows an example of an image captured by the camera 30 in this embodiment. According to the installation angle of the camera 30 described in FIG. 2 , the camera 30 can capture the image 100 when the car door 10 a (and the hall door) are in an open state. Specifically, as shown in FIG. 3 , the image 100 includes an area 101 on the elevator hall 20 side and an area 102 on the passenger car 10 side.

In addition, although not shown in FIG. 1 described above, a sill for guiding the opening and closing of the car door is provided on the side of the passenger car 10 . Similarly, on the hall 20 side, a sill for guiding the opening and closing of the hall door is provided. Therefore, in the image 100 shown in FIG. 3, the boundary portion between the area 101 on the side of the hall 20 and the area 102 on the side of the passenger car 10 includes a threshold provided on the side of the hall 20 (the hall side threshold ) 111 and a door sill (car side door sill) 112 provided on the passenger car 10 side.

In addition, a frame of the landing called a door frame is provided at the landing of the elevator hall 20 . The door frame consists of vertical frames on both sides of the entrance and a horizontal frame on the upper side. According to the above-mentioned installation angle of the camera 30 , the image 100 also includes a door frame (the vertical frames on both sides) 113 . In addition, since the camera 30 is installed under (a side plate portion of) the upper frame of the door frame 113 , the upper frame is not included in the image 100 .

In addition, when there is a user who wants to board the passenger car 10 in the hall 20, the user 200 is included in the area 101 on the hall 20 side in the image 100 as shown in FIG. 3 .

Returning again to FIG. 1 , the image processing device 40 is connected to the camera 30 . The image processing device 40 performs image processing (analysis processing) on the image (video) captured by the camera 30 in real time. The image processing device 40 is provided independently of the camera 30, for example. In addition, the image processing device 40 may be incorporated in the aforementioned car control device or the like.

The image processing device 40 has a function of detecting the situation of the elevator hall 20 by performing image processing on the image captured by the camera 30 . Specifically, the image processing device 40 can detect whether or not there is a user 200 who intends to board the passenger car 10 from the elevator hall 20 based on, for example, the area 101 on the elevator hall 20 side in the image 100 . Accordingly, for example, when it is detected that there is a user 200 who intends to board the passenger car 10, the open state of the car door 10a can be maintained through the associated operation of the image processing device 40 and the car control device. This control is performed until the user 200 finishes boarding the passenger car 10 .

In addition, the image processing device 40 can also detect, based on, for example, the area 102 on the side of the car 10 in the image 100 that the hand of the user riding on the car 10 is about to be pulled into the car door 10a during the door opening operation. The condition of the (door cover) (that is, the condition inside the passenger car 10). Accordingly, when it is detected that the user's hand is about to be pulled into the car door 10a, the image processing device 40 and the car control device can perform control such as interrupting the door opening operation through associated operations and the like.

Here, as described above, the image captured by the camera 30 includes a door frame, but the door frame often has a higher reflectance of light than other parts in the image. Therefore, when the camera 30 captures an image, the intensity of light reflected by the door frame may be strong, and this light may affect the image processing performed by the image processing device 40 as disturbance light. Specifically, although there is a user who intends to get on the passenger car 10 from the elevator hall 20, there may be a situation where the user cannot be detected. That is, there is a possibility that the detection accuracy of the situation of the elevator hall 20 may be lowered by the light reflected by the door frame.

Here, in the image processing device 40 , an area to be masked (an area including a door frame) is set as a setting value related to image processing. According to this, the light reflected by the door frame, which becomes disturbing light, can be blocked by the shielding process, so the influence of the disturbing light can be alleviated, and the detection accuracy of the situation of the elevator hall 20 can be improved.

Also, the area to be masked is set in advance based on a predetermined installation angle of the camera 30 .

However, the camera 30 is sometimes not installed at a predetermined installation angle. Specifically, since the camera 30 is installed manually by an operator, there is a possibility that the camera 30 is installed in a wrongly inclined state. In such a case, a misalignment occurs between the region to which the above-described masking process is performed and the region actually including the door frame in the image captured by the camera 30 . Accordingly, since the area including the door frame in the photographed image is not properly masked, the detection accuracy of the situation of the elevator hall 20 is reduced due to the influence of disturbance light (ie, light reflected by the door frame).

Furthermore, since the camera 30 is installed above the landing of the car 10 as described above, even when the car 10 itself is tilted, it may be similarly affected by disturbing light.

Therefore, the image processing device 40 according to the present embodiment has a setting value related to the above-mentioned image processing (which is masked) when the installation angle of the camera 30 is shifted (that is, the camera 30 is installed in a tilted state). A function to perform correction (automatic adjustment) (hereinafter referred to as a calibration function).

FIG. 4 is a block diagram showing the functional configuration of the image processing device 40 according to the present embodiment. Here, the functional configuration related to the above-mentioned calibration function will be mainly described.

As shown in FIG. 4 , the image processing device 40 includes a storage unit 41 , an image acquisition unit 42 , a reference site extraction unit 43 , a difference calculation unit 44 , a correction processing unit 45 , and a notification processing unit 46 .

The above-described setting values related to image processing are stored in the storage unit 41 . The setting values related to image processing include the area (coordinate value) where the door frame should be located as the area to be masked. In addition, in the storage unit 41 , a predetermined reference position (hereinafter, referred to as a reference position) for a reference point described later is stored in advance. The reference position is the position of the reference site in the photographed image when the camera 30 is accurately installed at a predetermined installation angle.

The image acquisition unit 42 acquires an image captured by the camera 30 (hereinafter referred to as a captured image).

The reference site extraction unit 43 extracts a reference site included in the photographed image acquired by the image acquisition unit 42 . The reference site is a site used as a reference for determining whether the installation angle of the camera 30 is shifted, and includes, for example, a hall-side sill and a car-side sill. In addition, the hall side door sill and the passenger car side door sill are often formed of, for example, alumina or stainless steel. Therefore, the hall-side door sill and the car-side door sill can be easily extracted from the photographed image compared with other parts.

The difference calculating unit 44 calculates the difference between the position of the reference site and the reference position when the position of the reference site extracted by the reference site extracting unit 43 (the position in the captured image) does not match the reference position stored in the storage unit 41 . difference. In addition, the difference calculated by the difference calculation part 44 is represented by a coordinate value etc., for example.

The correction processing unit 45 acquires setting values related to image processing stored in the storage unit 41 . The correction processing unit 45 corrects the acquired setting values related to image processing (for example, the area to be masked) based on the difference calculated by the difference calculation unit 44 . The correction processing unit 45 stores the corrected setting values related to image processing in the storage unit 41 . The corrected setting values related to the image processing are used in the image processing executed when the situation of the above-mentioned elevator hall 20 is detected.

Here, in this embodiment, by correcting the setting values related to the image processing by the correction processing unit 45, even when the installation angle of the camera 30 is shifted, the situation of the elevator hall 20 can be suppressed. decrease in detection accuracy. However, during operation of the elevator system, the installation angle of the camera 30 may be further shifted (that is, the camera 30 is tilted) due to, for example, a collision with the passenger car 10 or the camera 30 .

In such a case, the notification processing unit 46 notifies (sends) that the installation angle of the camera 30 has shifted to (the manager of) a monitoring center that monitors the operation status of the elevator system, for example. In addition, for the situation that the installation angle of the camera 30 is shifted, it can be determined according to whether the difference recalculated by the difference calculation part 44 is greater than The determination is made by the difference calculated by the difference calculation unit 44 when correcting the above-mentioned setting values related to the image processing.

Next, the processing procedure of the image processing device 40 in the calibration function in this embodiment will be described with reference to the flowchart of FIG. 5 . In addition, the processing shown in FIG. 5 is performed, for example, before operation of an elevator system.

First, the image acquisition unit 42 acquires an image (photographed image) captured by the camera 30 (step S1).

The reference point extracting unit 43 extracts, for example, the hall side threshold and the car side threshold (regions) from the captured image as reference points in the captured image acquired in step S1 (step S2 ).

Next, the difference calculation unit 44 acquires the reference position stored in the storage unit 41 . The difference calculation unit 44 determines whether or not the attachment angle of the camera 30 is shifted based on the acquired reference position (step S3 ). In this case, the difference calculation unit 44 executes a determination process based on whether or not the positions in the photographed images of the hall-side threshold and the car-side threshold (reference portion) extracted in step S2 coincide with the reference position. Specifically, when the positions in the photographed images of the hall-side threshold and the car-side threshold coincide with the reference position, it can be determined that the installation angle of the camera 30 has not shifted. On the other hand, when the positions in the photographed images of the hall-side threshold and the car-side threshold do not match the reference position, it can be determined that the installation angle of the camera 30 has shifted.

When it is determined that the mounting angle of the camera 30 is not shifted ("No" in step S3), since it is not necessary to correct the setting values related to image processing, the process ends.

On the other hand, when it is determined that the installation angle of the camera 30 is shifted (YES in step S3), the difference calculating unit 44 calculates the photographing angle of the hall-side threshold and the car-side threshold (reference position). The difference between the position in the image and the reference position (step S4).

Next, the correction processing unit 45 acquires the setting values related to image processing stored in the storage unit 41 . The correction processing unit 45 corrects the area to be masked (the area where the door frame should be located) included in the acquired setting values related to image processing based on the difference calculated in step S4 (step S5 ). In the following description, for the sake of convenience, the area to be masked included in the setting values related to image processing is referred to as a mask area.

Here, step S5 will be specifically described with reference to FIG. 6 and FIG. 7 . In addition, in FIG. 6 and FIG. 7, the same part as FIG. 3 mentioned above is attached|subjected and demonstrated the same code|symbol as FIG. Similarly, the same parts in FIG. 6 and FIG. 7 are also described with the same reference numerals attached to each other.

First, FIG. 6 shows an example of an image captured by the camera 30 in a state where the installation angle of the camera 30 is not shifted (the camera 30 is not tilted).

The reference position 400 shown in FIG. 6 represents the position of reference points (the hall-side sill 111 and the car-side sill 112 ) when the camera 30 is accurately installed at a predetermined installation angle.

In the image 310 shown in FIG. 6 , the positions of (the boundary between) the hall-side sill 111 and the car-side sill 112 coincide with the reference position 400 , so it can be determined that the installation angle of the camera 30 is not shifted.

Here, as shown in FIG. 6 , the area of the door frame 113 in such an image 310 coincides with the masked area 311 . That is, when the installation angle of the camera 30 is not shifted, the masking process can be appropriately performed on the area of the door frame 113 .

On the other hand, FIG. 7 shows an example of an image captured by the camera 30 in a state where the installation angle of the camera 30 is shifted (the camera 30 is tilted).

In the image 320 shown in FIG. 7 , the positions of (the boundary portion of) the hall-side sill 111 and the passenger car-side sill 112 do not match the reference position 400, so it can be determined that the installation angle of the camera 30 has shifted. .

Here, as shown in FIG. 7 , the area of the door frame 113 in such an image 320 does not coincide with the mask area 311 . The shielded area 311 is set as an area where the door frame 113 is located when the camera 30 is accurately installed at a predetermined installation angle. However, when the installation angle of the camera 30 is shifted, the area where the door frame 113 is located in the image 320 is shifted from the position of the shielded area 311 . That is, when the installation angle of the camera 30 deviates, the light reflected by the part of the door frame 113 that does not overlap with the shielded area 311 becomes disturbing light and affects it. There is a possibility that the detection accuracy of 20 situations will decrease.

Therefore, in step S5, the correction processing unit 45 uses the difference calculated in step S4 (that is, the coordinate value representing the inclination of the hall-side threshold 111 and the car-side threshold 112 relative to the reference position) , correct the position of the masked area 311 to be consistent with the area of the door frame 113 in the image 320 . In this case, the correction processing unit 45 may perform correction so as to deform the mask region 311 .

Returning again to FIG. 5 , the masked area corrected in step S5 (corrected masked area) is stored in the storage unit 41 (step S6 ).

By using such a corrected masking area, even in the case where the installation angle of the camera 30 is shifted, for example, as shown in FIG. The area of the door frame 113 is suitably shielded.

In addition, in FIG. 5 , the reference point has been described as the hall-side sill and the car-side sill, but the reference point may be one of the hall-side sill and the car-side sill. In addition, as long as the reference site is a site that can be extracted from a captured image, it may be a site other than the hall-side threshold and the car-side threshold.

In addition, in step S3 shown in FIG. 5 , as described above, it is determined whether the installation angle of the camera 30 is shifted based on whether the positions in the photographed images of the hall-side threshold and the passenger car-side threshold coincide with the reference position. . However, even if the installation angle of the camera 30 is shifted, as long as the shift (amount) is to a degree that does not affect the detection accuracy of the condition of the elevator hall 20, no correction and image processing may be performed. Composition of the relevant setpoints (mask area). That is, the process of step S3 may also be performed according to whether the difference between the positions in the photographed images of the hall-side threshold and the passenger car-side threshold and the reference position is within a predetermined range. If it is within the range, it is determined that the installation angle of the camera 30 has shifted.

As described above, in the present embodiment, the image captured by the camera 30 (photographed image) is acquired, and the difference between the position of the reference site included in the acquired captured image and the reference position (predetermined reference position) is calculated. , and correct the setting values related to the image processing according to the calculated difference. The image processing-related setting values include, for example, areas that are masked to reduce the influence of disturbing light on image processing. In addition, the area to which the masking process is performed includes, for example, the area where the door frame installed at the entrance of the elevator hall 20 is located.

In this embodiment, with such a configuration, even if the installation angle of the camera 30 (that is, the elevator hall 20 ) Even in the case where there is a deviation in the shooting angle of the camera, etc., the masking process can be appropriately performed on the area of the door frame. Accordingly, it is possible to suppress a decrease in the detection accuracy of the situation of the elevator hall 20 (or inside the passenger car 10 ) due to the deviation of the installation angle of the camera 30 .

In addition, in the present embodiment, the photographed image includes the elevator hall door 10a installed in the elevator hall 20 and the elevator hall 20 and the passenger car when the car door 10a installed in the passenger car 10 is in an open state. The boundary portion of the car 10 , the reference portion includes at least one of the hall-side sill and the passenger car-side sill. In this embodiment, with such a configuration, the hall-side threshold and the car-side threshold can be easily extracted from the photographed image, so the position of the reference point can be accurately compared with the reference position to detect (determine) ) The offset of the installation angle of the camera 30.

In addition, in this embodiment, the camera 30 is installed on the side plate portion above the entrance and exit of the passenger car, so that both the waiting hall 20 and the inside of the passenger car 10 can be taken as the imaging range. . Thus, as described above, it is possible to detect the situation of the elevator hall 20 (whether there is a user who intends to get on the passenger car 10 from the elevator hall 20) and the situation in the passenger car 10 (the user's hands or not). etc. will be pulled into the car door 10a) etc.

Furthermore, in this embodiment, after correcting the setting values related to image processing, the difference between the position of the reference part included in the image captured by the camera 30 and the reference position is further calculated. If the difference is larger than the difference calculated when correcting the set values related to image processing, the administrator is notified of an abnormality (that is, when the installation angle of the camera 30 is further shifted). In the present embodiment, with such a configuration, for example, when the installation angle of the camera 30 is further misaligned after correcting the setting values related to image processing, the administrator can be urged to correct the misalignment.

In addition, in the present embodiment, the masked area (the area to be masked) is corrected as the setting value related to the image processing. set value.

Specifically, in the image processing device 40 according to the present embodiment, for example, when the car door 10a is in an open state, in order to detect the situation of the elevator hall 20, the area on the elevator hall 20 side is targeted. Exposure adjustment of the camera 30 is performed. In addition, for example, when the car door 10a is in the closed state or in the case of opening the door, in order to detect the situation in the car 10, the area on the side of the car 10 is used as a target and the camera 30 exposure adjustment. In this case, the area (location) on the side of the elevator hall 20 and the area (location) on the side of the passenger car 10 that are subject to exposure adjustment is set in advance as a setting value related to image processing. However, as described above, when the installation angle of the camera 30 is shifted, the set area may deviate from the area on the elevator hall 20 side or the area on the passenger car 10 side where exposure adjustment should be performed. shift, and proper exposure adjustments cannot be made. Therefore, in the present embodiment, a configuration may be adopted in which an area to be subject to exposure adjustment when an image is captured by the camera 30 is corrected based on a difference between the position of the reference part and the reference position.

Although some 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 embodiments can be implemented in other various forms, and various omissions, substitutions, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and spirit of the invention, and are also included in the invention described in the claims and their equivalents.

Claims (7)

1. a kind of image processing apparatus, it takes the situation of the hall of cage to detect user, and for by The image including the hall that video camera is photographed performs image procossing, and described image processing unit is characterised by, has It is standby：

Acquisition unit, it obtains the image photographed by the video camera；

Computing unit, the position at its benchmark position included in acquired described image is advance with being directed to the benchmark position In the case that the position as benchmark of determination is inconsistent, the difference of the position and the position as the benchmark at the benchmark position is calculated Point；And

Unit is corrected, its difference calculated according to corrects the setting value relevant with described image processing.

2. image processing apparatus according to claim 1, it is characterised in that

The setting value relevant with described image processing includes being carried out the influence that described image is handled to reduce interference light The region of shielding processing.

3. image processing apparatus according to claim 2, it is characterised in that

The region for being carried out shielding processing includes the region that the doorframe for the stopping port for being arranged at the hall is located at.

4. image processing apparatus according to claim 1, it is characterised in that

Exposure of the being used as when setting value relevant with described image processing includes shooting image by the video camera is adjusted The region of whole object.

5. image processing apparatus according to claim 1, it is characterised in that

The image photographed by the video camera is included in the hall door for being arranged at the hall and is arranged at described multiply The boundary member of the hall and the cage in the case of being in open mode with the car door of car,

The benchmark position is including for the hall side door sill for guiding the opening and closing of the hall door and setting and in order to draw At least one party in the cage side door sill led the opening and closing of the car door and set.

6. image processing apparatus according to claim 5, it is characterised in that

The video camera is arranged at the top of the stopping port of the cage.

7. image processing apparatus according to claim 1, it is characterised in that

It is also equipped with notification unit,

The computing unit is further calculated in the image obtained after the correction setting value relevant with described image processing The position at included benchmark position and the difference of the predetermined position as benchmark,

The notification unit is in the case where the difference further calculated is more than the difference calculated, to management Person notifies abnormal.