Automated monitor arm for object recognition

Abstract

The present invention relates to an object recognition fully automatic monitor arm, and more specifically, to an object recognition fully automatic monitor arm for automatically adjusting the position of a monitor according to the user. In addition, it is characterized by including an arm for mounting a monitor, a drive unit for rotating the arm from a table or adjusting its height, a camera for photographing a user, and a control unit for controlling the position of the monitor by controlling the drive unit based on an image captured through the camera.

Description

Technology Field

[0001] The present invention relates to an object recognition fully automatic monitor arm, and more specifically, to an object recognition fully automatic monitor arm for automatically adjusting the position of a monitor according to the user. Background Technology

[0003] With the rapid advancement of digitalization, computer-centered work is gradually increasing in the workplace and daily life.

[0005] Computer-assisted work is performed by the worker sitting in a chair, looking at a monitor placed on a desk, and using a keyboard and mouse.

[0007] In particular, office workers spend most of their working hours using computers, which places a significant physical strain on them, and there are many cases where they complain of musculoskeletal discomfort such as eye strain and neck and shoulder pain.

[0009] Accordingly, the use of monitor arms to adjust the position of monitors according to the operator is increasing; however, existing monitor arm solutions had the problem that it was difficult to maintain the optimal position through manual operation and repetitive adjustments were cumbersome.

[0011] To solve these problems, electric monitor arms are being developed, such as in Korean Patent Publication No. 10-2018-0124616 "Electric Height-Adjustable Monitor Stand," but there was still the inconvenience of having to control the position of the monitor. Prior art literature

[0013] Korean Published Patent No. 10-2018-0124616 "Electric height-adjustable monitor stand" The problem to be solved

[0014] The objective of the present invention is to provide an object recognition fully automatic monitor arm for automatically controlling the position of a monitor according to the user's posture and position, which is devised to solve the problems described above. means of solving the problem

[0016] To achieve the above objective, the object recognition fully automatic monitor arm according to the present invention is characterized by comprising an arm for mounting a monitor, a driving unit for rotating the arm from a table or adjusting its height, a camera for photographing a user, and a control unit for controlling the position of the monitor by controlling the driving unit based on an image captured through the camera.

[0017] In addition, the above drive unit is characterized by including a rotary drive unit for horizontal rotation and a vertical drive unit for moving the mounted monitor in the up and down directions.

[0018] In addition, the control unit is characterized by including a data collection step for acquiring captured image data, an extraction step for extracting an object area from the image data using a YOLO model, a coordinate storage step for generating and storing coordinate data of the extracted object area, a calculation step for forming correction position data of a monitor corresponding to the stored coordinate data of the object area, and a control step for controlling the driving unit based on the calculated correction position data to move the monitor to the correction position.

[0019] In addition, the control unit comprises a data collection step for acquiring captured image data, an extraction step for extracting an object area from the image data using a YOLO model, a coordinate storage step for generating and storing coordinate data of the extracted object area, a learning step for learning correct coordinate data when in a correct posture based on changes in the stored object area coordinate data, a calculation step for forming correct position data of a monitor corresponding to the learned correct coordinate data, and a control step for controlling the driving unit based on the calculated correct position data to move the monitor to the correct position. Effects of the invention

[0021] As described above, the object recognition fully automatic monitor arm according to the present invention has the effect of automatically controlling the position of the monitor according to the user's posture and position by tracking the user through real-time object recognition. Brief explanation of the drawing

[0023] FIG. 1 is a flowchart illustrating the operation of an object recognition fully automatic monitor arm according to the present invention in sequence. FIG. 2 is a diagram illustrating the operation of an object recognition fully automatic monitor arm according to the present invention in a simplified diagram. FIG. 3 is a drawing illustrating an object recognition fully automatic monitor arm according to the present invention. FIG. 4 is a drawing illustrating the arm and driving unit of an object recognition fully automatic monitor arm according to the present invention. Specific details for implementing the invention

[0024] Specific structural or functional descriptions of embodiments according to the concept of the present invention disclosed herein are provided merely for the purpose of explaining embodiments according to the concept of the present invention, and embodiments according to the concept of the present invention may be implemented in various forms and are not limited to the embodiments described herein.

[0025] Embodiments according to the concept of the present invention may be subject to various modifications and may take various forms; therefore, embodiments are illustrated in the drawings and described in detail in this specification. However, this is not intended to limit the embodiments according to the concept of the present invention to specific disclosed forms, and includes all modifications, equivalents, or substitutions that fall within the spirit and scope of the present invention.

[0027] Hereinafter, preferred embodiments of the present invention will be described with reference to the attached drawings.

[0029] FIG. 1 is a flowchart illustrating the operation of an object recognition fully automatic monitor arm according to the present invention in sequence, FIG. 2 is a diagram illustrating the operation of an object recognition fully automatic monitor arm according to the present invention in a simplified diagram, FIG. 3 is a diagram illustrating an object recognition fully automatic monitor arm according to the present invention, and FIG. 4 is a diagram illustrating the arm and driving unit of an object recognition fully automatic monitor arm according to the present invention.

[0031] As illustrated in FIGS. 1 to 4, the object recognition fully automatic monitor arm according to the present invention comprises an arm (1) for mounting a monitor (4), a driving unit (2) for rotating the arm (1) from a table or adjusting its height, a camera (3) for photographing a user, and a control unit (not shown) for controlling the position of the monitor (4) by controlling the driving unit (2) based on an image captured through the camera (3).

[0033] At this time, the arm (1) is configured to rotate about the Z-axis to rotate the monitor (4) from the table in a horizontal (left, right) direction, and is configured to be able to move in a vertical (up, down) direction or in a horizontal (forward, backward) direction to adjust the height of the monitor (4).

[0035] Additionally, the drive unit (2) is configured to include a rotary drive unit (21) for rotating the arm (1) horizontally to rotate the position of the monitor (4) in a horizontal direction, and a vertical drive unit (22) for moving the arm (1) vertically to move the position of the mounted monitor (4) up and down.

[0037] Additionally, the above-mentioned drive unit (2) may be implemented to further include a horizontal drive unit (not shown) for moving the arm (1) in the horizontal (forward, backward) direction.

[0039] The object recognition fully automatic monitor arm according to the present invention, having such a configuration, is controlled through the control unit via the following operation.

[0041] More specifically, the method comprises a data collection step (S1) for acquiring image data captured through the camera (3), an extraction step (S2) for extracting an object area (100) from the image data using a YOLO model, a coordinate storage step (S3) for generating and storing coordinate data of the extracted object area (100), a calculation step (S4) for forming correction position data of a monitor corresponding to the stored coordinate data of the object area (100), and a control step (S5) for controlling the driving unit (2) based on the calculated correction position data to move the monitor (4) to the correction position.

[0043] In other words, the user is photographed with a camera (3) through the data collection step (S1), and an object area (100) regarding the user is extracted from the image in the extraction step (S2), and then coordinate data of the extracted object area (100) is generated in the coordinate storage step (S3).

[0044] Afterwards, correction position data for moving the monitor (4) is formed in correspondence with the coordinate data generated in the calculation step (S4), and the position of the monitor (4) is adjusted to the user's position through the control step (S5).

[0046] This may be executed according to user control, or if the coordinate data for the object area (100) regarding the user changes beyond a preset range (virtual bounding box), it may be executed again from the data collection step (S1).

[0048] Alternatively, the position of the monitor (4) can be continuously adjusted to the user at all times by running it repeatedly at pre-set time intervals or by immediately correcting the position of the monitor (4) in real time according to the user's posture change or movement.

[0050] In addition, the extraction step (S2) is performed using a high-performance deep learning-based object recognition algorithm, and more specifically, image information is received in real time through a camera (3), and the object region (100) can be extracted from the image data using a SORT (Simple Online Realtime Tracking) algorithm that tracks the movement of an object over time together with a YOLO model trained on a Convolutional Neural Network (CNN).

[0052] In another embodiment, the position of the monitor (4) may be adjusted so that the user can view the monitor (4) in a more correct posture.

[0054] More specifically, the control unit may be configured to include a data collection step (S1) for acquiring captured image data, an extraction step (S2) for extracting an object area (100) from the image data using a YOLO model, a coordinate storage step (S3) for generating and storing coordinate data of the extracted object area (100), a learning step (S4-1) for learning correct coordinate data when in a correct posture based on changes in the stored coordinate data of the object area (100), a calculation step (S4-2) for forming correct position data of a monitor corresponding to the learned correct coordinate data, and a control step (S5) for controlling the driving unit (2) based on the calculated correct position data to move the monitor (4) to the correct position.

[0056] That is, the data collection step (S1), extraction step (S2), and coordinate storage step (S3) are repeatedly performed, and based on the stored coordinate data, the correct coordinate data corresponding to when the user is sitting in a correct posture is learned through the learning step (S4-1). Based on this, the monitor (4) is not moved to a position corresponding to the user's current posture, but rather the monitor (4) is moved to a position corresponding to when the user is in a correct posture.

[0058] Alternatively, by controlling the drive unit (2) from the monitor position corresponding to the current posture to the monitor position corresponding to the correct posture, the monitor (4) can be moved gradually (stepwise) or at a preset speed at preset time intervals, thereby inducing a change in the user's posture so that the user changes from a bad current posture to a correct posture by repeating the process of the user looking at the monitor (4) at the changed position through the change in the position of the monitor (4).

[0060] As described above, although the present invention has been described with reference to preferred embodiments with reference to the accompanying drawings, it is evident to those skilled in the art that many obvious variations are possible from this description without departing from the scope of the invention. Accordingly, the scope of the invention should be interpreted by the claims described to include examples of such many variations. Explanation of the symbols

[0062] 1 : Cancer 2 : Driving unit 3 : Camera 4 : Monitor

Claims

Claim 1 An object recognition fully automatic monitor arm characterized by comprising: an arm for mounting a monitor; a driving unit for rotating the arm from a table or adjusting its height; a camera for capturing a user; and a control unit for controlling the position of the monitor by controlling the driving unit based on an image captured through the camera. Claim 2 An object recognition fully automatic monitor arm, characterized in that, in claim 1, the driving unit comprises a rotary driving unit for rotation in a horizontal direction and a vertical driving unit for moving a mounted monitor in an up and down direction. Claim 3 The object recognition fully automatic monitor arm according to claim 1, wherein the control unit comprises: a data collection step for acquiring captured image data; an extraction step for extracting an object region from the image data using a YOLO model; a coordinate storage step for generating and storing coordinate data of the extracted object region; a calculation step for forming correction position data of a monitor corresponding to the stored coordinate data of the object region; and a control step for controlling the driving unit based on the calculated correction position data to move the monitor to the correction position. Claim 4 In claim 1, the control unit comprises: a data collection step for acquiring captured image data; an extraction step for extracting an object region from the image data using a YOLO model; a coordinate storage step for generating and storing coordinate data of the extracted object region; a learning step for learning correct coordinate data when in a correct posture based on changes in the stored object region coordinate data; a calculation step for forming correct position data of a monitor corresponding to the learned correct coordinate data; and a control step for controlling the drive unit based on the calculated correct position data to move the monitor to the correct position.

Images