KR20130078496A - Apparatus and method for controlling electric boards using multiple hand shape detection and tracking - Google Patents

Abstract

PURPOSE: An electronic board remote control apparatus by detecting and tracing various hand poses and a method thereof are provided to remotely control an electronic board without touching the board by detecting a hand and tracking a hand pose. CONSTITUTION: A camera image acquisition unit (100) photographs a user who is around an electronic board. A Region of Interest (ROI) setup unit (200) sets up the image region, in which a hand is located, out of images photographed by the camera image acquisition unit as an ROI. A hand pose detection unit (300) detects a hand pose using a hierarchical image analysis of the ROI. A control event generation unit (400) generates a designated control event corresponding to the hand pose in order to control the electronic board. An electronic board control unit (500) remotely controls the electronic board based on the control event. [Reference numerals] (100) Camera image acquisition unit; (200) Region of Interest (ROI) setup unit; (300) Hand pose detection unit; (400) Control event generation unit; (500) Electronic board control unit

Description

Remote electronic blackboard control device and method thereof based on various hand detection and tracking {APPARATUS AND METHOD FOR CONTROLLING ELECTRIC BOARDS USING MULTIPLE HAND SHAPE DETECTION AND TRACKING}

The present invention provides a remote control function to the electronic blackboard control provided only by the existing touch method through the detection and tracking of various hand shapes, the instructor who delivers learning can control the electronic blackboard in various spaces. It relates to an apparatus and a method for providing the same.

Recently, the electronic blackboard market is expanding greatly based on national support for improving educational facilities and demand for constructing smart education environment combined with IT. The electronic board-related market forms a large market of more than 1 trillion won, including the content market, and the market is expected to grow further. The core technology of this copyboard is how to provide the user with effective interaction on the large screen. Most electronic blackboard products apply touch technology to solve these problems, and recently, multi-touch technology has been developed and applied.

Research on various user interfaces using gestures is being conducted, and in the case of an electronic blackboard based on touch recognition technology in the case of a conventional electronic blackboard, when an actual touch does not occur, the user cannot receive input from the user. Thus, suggestions are made for interaction with the user (or instructor). Various proposals have been made to remove this limitation, but in most cases additional tools or devices are required, which makes the application difficult.

The present invention is to provide a system that enables the user to interact with the electronic blackboard without any other tools by detecting and tracking various user shapes using a camera mounted on the electronic blackboard.

The present invention prioritizes various hand shapes of a user in a general camera or an infrared camera, and sets a region of interest (ROI) to effectively detect hand shapes in a continuous image. In addition, the present invention provides a method for detecting the motion through continuous tracking based on the set ROI and providing an effective interface with the copyboard based on the motion and hand shape.

According to an aspect of the present invention, as a remote electronic blackboard control device,

A camera image acquisition unit for photographing a user who is remote from the copyboard;

A region of interest setting unit configured to set a region where a user's hand is located among regions of the image acquired by the camera image acquisition unit as a region of interest (ROI);

A hand shape detector configured to detect a hand shape taken by a user through hierarchical image analysis of the ROI; And

Control event generation unit for generating a predetermined control event for the control of the electronic blackboard corresponding to the hand shape detected by the hand shape detection unit

There is provided a gesture-based remote electronic blackboard control apparatus comprising a.

The electronic board controller may further include an electronic blackboard controller configured to perform remote control on the electronic blackboard based on the generated control event.

In one embodiment, the camera image acquisition unit,

By invoking a calibration routine, the user checks at least one of the size and shape of the copyboard when the user looks at the copyboard at a distance from his or her location. Corresponding scale values can be set automatically.

In one embodiment, the camera image acquisition unit,

Analyzing the image related to a predetermined calibration gesture taken by the user in the course of performing the calibration routine,

When the user looks at the copyboard by calculating one or more of the diagonal length of the rectangular copyboard, the size of the angle formed when the two diagonal lines intersect, the position of each vertex, and the length of at least three of the four sides. At least one of the size and shape of the copyboard of the can be checked, and based on the result of the check, the scale value with respect to the horizontal axis and the vertical axis can be set to a fixed scale value or a scale function in which the scale value varies depending on the position.

In one embodiment, the camera image acquisition unit,

Pan-Tilt-Zoom (PTZ) control is included so that the user's torso can be rotated or zoomed in at least one of the up, down, left and right directions with respect to the camera so that the user's upper body enters the center of the image. You can execute Zoom Out.

In one embodiment, the region of interest setting unit,

Set the region of interest based on the detected position of the hand, or set the region of movement as the region of interest using the difference image comparing the current image with the previous image acquired by the camera image acquisition unit, or skin color The sample may be used to set a region where skin color is detected as the region of interest.

In one embodiment, the hand detector comprises an individual hand detector for detecting a predetermined individual hand shape, wherein the predetermined hand shape comprises a moving cursor hand shape, an executing hand shape, a selection moving hand. Contains at least one of the shapes-,

The predetermined individual hand gestures may be detected through hierarchical steps for detecting the individual hand gestures in an order based on a preset priority.

In one embodiment, the hand detection unit,

Only when the execution hand gesture is subsequently detected after a predetermined preceding gesture is detected, it is possible to determine that the detection of a valid execution hand gesture is made.

In one embodiment, the hand detection unit,

The position change in the image generated according to the hand shape is calculated based on the detected hand shape, and the position change in the image space is converted into the movement on the actual copyboard by reflecting the set scale value in the calculated position change amount. It can be done.

According to another aspect of the invention, as a method for remotely controlling the electronic blackboard,

By using an image of a user remote from the copyboard, at least one of the size and shape of the copyboard when the user looks at the copyboard at his / her location is checked, and the user's gesture on the viewing point of the user Setting a calibration scale value to be applied correspondingly;

Setting an area in which the user's hand is located as a region of interest using an image of a user remote from the copyboard;

Detecting a hand shape taken by a user through hierarchical image analysis of the ROI;

Generating a predetermined control event for controlling the electronic blackboard corresponding to the detected hand shape; And

There is provided an electronic blackboard remote control method comprising performing a remote control on the electronic blackboard based on the generated control event.

According to an embodiment of the present invention, the control on the electronic blackboard can be controlled remotely without touching by hand detection and hand movement tracking.

In addition, according to the embodiment of the present invention, it is possible to reduce the time required for execution to the minimum through the ROI setting and hierarchical hand shape detection.

In addition, according to an embodiment of the present invention, through the detection of various hand shapes, the menu or icon can be selected through the movement of the cursor, the menu or icon can be executed or moved, and the selected points can be selected through various hand shapes. ZOOM IN / OUT can be performed.

1 is a block diagram schematically illustrating various hand-based electronic blackboard control.
2 is a flow chart for a PTZ control routine for resetting a target area based on calibration and upper body detection of a person with respect to an image obtained from a camera.
3 is an example of a calibration gesture for giving a size of a selection area for scaling setting.
4 is a flowchart illustrating a process of setting a region of interest from a given image.
5 is a flowchart illustrating a process of hierarchically recognizing various hand shapes from an ROI.
6 is a flowchart showing the occurrence of an event based on the detected hand shape and the amount of movement.
7 is an example of a Harr-like feature extraction used for feature extraction for hand detection.

The present invention may be variously modified and have various embodiments, and specific embodiments will be illustrated in the drawings and described in detail with reference to the accompanying drawings. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings. In addition, numerals (e.g., first, second, etc.) used in the description of the present invention are merely an identifier for distinguishing one component from another.

In addition, in this specification, when one component is referred to as "connected", "connected", "transmitted", etc. with another component, the one component is directly connected to the other component, or directly connected It may be directly transmitted, but it is to be understood that unless otherwise stated, it may be connected, connected or transmitted via another component in the middle.

Hereinafter, with reference to the accompanying drawings will be described embodiments of the present invention; Although the present invention focuses on an interface with an electronic blackboard, the present invention may be applied to a method of using a gesture in various information devices providing an interface using a mouse, even if the electronic blackboard is not.

1 is a hand-based electronic blackboard control block diagram.

Referring to FIG. 1, a remote electronic blackboard control apparatus based on hand shape detection may include a camera image acquisition unit 100, an ROI setting unit 200, a hierarchical hand shape detection unit 300, and a hand recognition recognition based control event. It includes a generation unit 400 and the event-based electronic blackboard control unit 500. Hereinafter, each component will be described in detail with reference to FIGS. 2 to 7.

2 is a flowchart of a PTZ control routine for resetting a target area based on calibration and detection of a human upper body with respect to an image obtained from a camera. 3 is an example of a calibration gesture for giving a size of a selection area for scaling setting. Hereinafter, the functions and roles of the camera image acquisition unit 100 of FIG. 1 will be described with reference to FIGS. 2 and 3.

The camera image acquisition unit 100 may include a camera module (not shown), and when referring to FIG. 2, an image may be acquired through the camera module (see step S105).

Thereafter, the camera image acquisition unit 100 calls a calibration routine according to step S110. The calibration routine is a process of checking the size of the copyboard in the distance based on the movement of the viewpoint at the current position of the user, and also setting the general gesture movement of the user through the minimum gesture.

When the calibration routine is called, the user takes a specific gesture for such calibration, and the camera image acquisition unit 100 performs calibration by analyzing the image obtained thereby (see step S120).

Referring to step S120 as follows.

When the user takes a gesture (movement) starting from the upper left to the lower right at the position to be set as shown in FIG. 3, for example, at the position of the user who is remote from the copyboard based on the gesture. You can check the size of the copyboard when the user looks at it. For example, assuming that the camera module is installed at an arbitrary position of the edge of the copyboard (eg, the upper edge of the edge), a plurality of image frames in which the user's gestures are continuously captured by the camera module are displayed. Once acquired, the camera image acquisition unit 100 can check both end points of the diagonal length and the diagonal direction on the rectangular electronic blackboard when the user looks by tracking the corresponding gestures in the plurality of image frames. The overall shape and size of the copyboard can be confirmed.

Of course, in the above example, it is assumed that the user is looking at the front of the copyboard, the camera module is installed in a position such as the center of the upper edge of the copyboard, but may be installed in various locations in addition to the camera module Of course, it is not necessary to be installed (fixed or attached) to a specific part of the copyboard. That is, even if the camera module is installed in a position separated from the copyboard, it may be disposed / installed only at a position capable of capturing a user's gesture. However, if the user captures the user's gesture in front of the copyboard from the side, the process of calculating the size of the copyboard may be a little more complicated. However, this is only related to software processing. For there will be no great crowd.

However, hereinafter, for the convenience of understanding and explanation of the present invention, the camera module is installed at the edge of the copyboard, and the user will be described on the assumption that the user takes various gestures while viewing the copyboard from the front. .

In addition, the above assumes that the user makes a diagonal gesture from the upper left corner to the lower right corner of the copyboard to the user, but on the contrary, a diagonal gesture from the upper right corner to the lower left corner, Of course, various gestures may be used, such as taking a "-" or "b" -shaped gesture from the upper left corner to the lower right corner. However, this is because the copyboard is not tilted so that the user generally has a rectangular or square shape, and the user looks directly at the copyboard so that the original shape of the copyboard (for example, rectangular or Square shape) may be useful in situations where it may remain.

That is, if the copyboard is tilted, the copyboard has a trapezoidal shape that is shorter in length than the side facing one of the upper side, the lower side, the left side, and the right side. It may have a short rectangular shape. In addition, if the user looks at the copyboard in the lateral direction it may have a parallelogram shape. In order to take into account the difference between the user's viewpoint and / or the tilt of the copyboard as in the above case, as shown in FIG. 3, the gesture is taken from the upper left corner of the copyboard to the lower right corner diagonally. Subsequently, the size and shape of the copyboard can be confirmed through a gesture ranging from the lower right corner (vertex) to the diagonal direction from the upper right corner to the lower left corner. Of course, as described above, it is obvious that various gestures different from those in the above example may be used. This includes the diagonal length of the square shape, the size of the angle formed by the two diagonal lines, the position of each vertex (which may correspond to the position of the inflection point that suddenly changes when the gesture is tracked), and the four sides. This is because any gesture may be used as long as a gesture capable of calculating a combination of at least one of at least three sides (which is possible by calculating a distance between each vertex) and the like. In the present specification, such gestures will be referred to as calibration gestures.

According to the calibration gesture, the x-axis scale element and the y-axis scale element related to the movement of the user at the current position of the user can be set. Here, the x-axis scale value is named sx, and the y-axis scale value is named sy. This can be done by considering the ratio of the original resolution of a given copyboard to the measured image space. And if you want to get more accurate scale value even when the size ratio of the top and bottom is not the same as the tilted copyboard, the scale value of the y axis is different from each row of the y axis with the function of sy (yi). Can be represented in the same format.

In addition, when PTZ (Pan-Tilt-Zoom) control of the camera module is possible, the upper body of the instructor or the user is detected and the upper body is centered in the image based on the detected upper body to track optimal hand movements. The camera module can be rotated up / down / left / right or / and zoomed in to zoom out (see steps S125 and S130). For example, when the upper body size detected in the image is a constant ratio of the height of the entire image, for example, less than 1/4, the ZOOM IN function can be used to track the motion in the larger image. On the contrary, if the user comes out of the camera too large, for example, if it is 1/2 or more, the movement of the hand does not leave the image through the ZOOM OUT. Of course, it is also possible to adjust the camera module by rotating the camera module up / down / left / right.

In this way, when calibration initialization or PTZ control occurs, an image after movement is provided for new hand detection. That is, the initialization and PTZ control of the foregoing process may be repeated. For example, when ZOOM IN or ZOOM OUT occurs, the scale elements sx and sy are modified accordingly accordingly.

Although described above with reference to the flowchart of FIG. 2, step S125 and step S130 in FIG. 2 may be performed before step S110 and step S120 according to a situation or a design implementation method. That is, based on the user photographed image obtained through the camera image acquisition unit 100, first check whether the upper body of the user is located at the center of the camera (that is, the center of the image), and when the above conditions are completed, the calibration process is performed after It will be apparent to those skilled in the art that the manner of performing the present invention can be easily understood without a detailed description.

When the calibration initialization process as described above is completed, the acquired image is transferred to the ROI setting unit 200 as an image for hand detection (see step S150).

4 is a flowchart illustrating a process of setting a region of interest from a given image. Hereinafter, the function and role of the ROI setting unit 200 of FIG. 1 will be described in detail with reference to FIG. 4.

Instead of sending the entire image to the hierarchical hand shape detection unit 300, the image obtained by the camera image acquisition unit 100 performs hand detection only in an area where an actual hand may exist through the ROI setting unit 200. By doing so, the processing speed is improved. As shown in FIG. 4, this ROI is used as the ROI set through the hand detection if it was previously detected (see steps S205 and S210). In step S215, a region of interest is set through the following process.

Unlike when a region of interest has been set up and is being processed (that is, the user is continuously executing the remote control of the copyboard), when a region of interest must be newly set up (that is, the user starts remote control of the copyboard) To compare the current image with the previous image based on the previous image and the current image acquired by the camera image acquisition unit 100, and extract the motion through the difference image region for this (step S215). If there is motion, the ROI may be set based on the motion area (that is, the difference image area) (see step S220). If no movement is found, a region having a skin color sample can be set as the region of interest among regions where the skin color is detected (see steps S225 and S230). If the detection is not possible in this way, the entire input image may be set as the ROI (see step S240). However, if the ROI cannot be detected through the above-described steps S205, S215, and S225, the entire process of comparing the input image with the next input image may be performed without detecting the entire input image as the ROI as in step S240. Of course.

The detection subject region (ie, the region of interest) set as described above is used for the next hand detection, which is transmitted to the hierarchical hand shape detecting unit 300 (see step S250).

5 is a flowchart illustrating a process of hierarchically recognizing various hand shapes from an ROI. Hereinafter, the function and role of the hierarchical hand shape detection unit 300 of FIG. 1 will be described in detail with reference to FIG. 5.

The hierarchical hand detection unit 300 does not detect several types of hands at the same time even when detecting hand types by step as shown in FIG. The reaction rate may be greatly reduced.) The hierarchical structure of the shape of the hand to be detected is detected, so that the detection is performed based on the hierarchical structure.

At this time, each hand shape can be used by learning an individual hand detector (not shown), the individual hand detector is a cascade Adaboost-based hand detector or a variant or SVM-based hand shape Detectors can be used. Here, a case of performing a hand gesture recognition based on cascade Adboost will be described, for example. A Harr-like feature value as shown in FIG. 7 is extracted, and a cascade Adapst learner is trained based on this. For this learning, hundreds to thousands of positive and negative images of each hand shape are collected and used for learning. This results in a learning system that can be individually recognized for each hand shape. Based on this, in order to detect the hand shape of the individual image after the learning, each hand shape detector detects the learned hand shape by searching the entire window with respect to the set window size and the size of a predetermined multiple thereof. In this case, an integral image is used to effectively calculate the harr-like feature value. The learning of the desired number of hands is done individually, and if you want to detect n individual hands, the detection speed is n times slower than one hand. Therefore, instead of detecting all the hand shapes at the same time, it has a hierarchical structure and detects the most commonly used hand shapes, and in the case of some hand types, it detects only when certain conditions are satisfied.

For example, if the most commonly used cursor movement hand shape is defined as a flat hand shape (ie, the shape of a hand), the single hand shape is detected by the cascade ad-boost method, and when detected, the hand motion is controlled. To transfer the change in position (see step S315 and step S320). This may be started by invoking a cursor movement hand detection routine for an image of the ROI transmitted from the ROI setting unit 200 (see steps S305 and S310).

If the hand shape is not detected, the execution hand shape that can be commonly used for a function such as double clicking of a mouse is detected (see step S330). The execution hand shape may set a hand shape such as a fist. It may be required to have a motion gesture for execution before detecting the execution hand shape, and accordingly, the execution gesture may be recognized only when a stroke has occurred (see step S325). If there is no stroke or the execution hand is not detected, the hand shape for the selection movement is detected, and if this hand shape is detected, the position change amount according to the hand movement is transmitted (see steps S340 and S350). . An example of a selective move hand shape may use a scissors hand shape.

In addition, in the case of the transfer of the position change amount, the change amount for the actual control is calculated by multiplying the scale value obtained in the previous calibration process or preset in the initialization process. That is, in the real image, if the change by Δx occurs, the value transmitted for the actual control is the change of the position (coordinate) by the value multiplied by sx (x-axis scale value) and Δx (x-axis displacement in the image). Will be Similarly, for Δy, the position change is made as much as the product of sy (y-axis scale value) and Δy (y-axis displacement in the image). In addition, by setting a gesture, an additional gesture, for example, a gesture for ZOOM IN / OUT may be set to control the zoom IN / OUT based on the movement of the hand based on this (see step S380). As described above, recognition for all gestures for the set hand shapes is performed step by step, and the detected hand shape and / or position change amount is transmitted to the hand recognition recognition control event generating unit 400 of FIG. 1 (step S360). , S370, S390]. Or, if no set hand shape is detected, a NO ACTION message is transmitted (see step S390).

6 is a flowchart showing the occurrence of an event based on the detected hand shape and the amount of movement. Hereinafter, the function and role of the hand recognition based control event generator 400 of FIG. 1 will be described in detail with reference to FIG. 6.

The amount of change in the detected hand shape and movement generates a control event according to each hand shape in the hand recognition recognition control event generation unit 400. 6 shows an example of the generation of such a hand gesture.

Referring to FIG. 6, when the hand shape detected by the hierarchical hand shape detection unit 300 is a cursor movement hand shape, the cursor is moved on the copyboard based on the detected hand shape and the movement amount (ie, the position change amount). Is converted to an instruction used to perform the operation (see steps S405, S415, S420). In addition, when the execution hand gesture is detected, a command corresponding to the menu or icon selected by the cursor of the previous step is executed (see steps S425 and S435). In the case of the selection moving hand shape, a command for moving the position of the selected object is generated (see steps S440 and S445), and a corresponding command is generated for an additional hand shape (see step S450).

In the command generation step, a hand diagram that must occur sequentially according to the operation is configured into a state diagram based on the possible scenarios, and the malfunction caused by the hand shape misdetection is reduced based on this. Command generation can be performed by the method of message transfer or by direct driving of the corresponding copyboard program.

The event-based copyboard control unit 500 of FIG. 1 receives a command generated by each hand gesture from the hand-shaped recognition-based control event generation unit 400 according to step S460 of FIG. 6, and the command received as described above. Based on the actual program. This part is divided from the control event generator because there is a difference in the specification of the copyboard and the API provided depending on the copyboard manufacturer. Therefore, the electronic blackboard controller 500 and the control event generator 400 may be merged according to the design implementation.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those of ordinary skill in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the following claims And changes may be made without departing from the spirit and scope of the invention.

Claims (12)

Remote copyboard control device,
A camera image acquisition unit for photographing a user who is remote from the copyboard;
A region of interest setting unit configured to set a region where a user's hand is located among regions of the image acquired by the camera image acquisition unit as a region of interest (ROI);
A hand shape detector configured to detect a hand shape taken by a user through hierarchical image analysis of the ROI; And
Control event generation unit for generating a predetermined control event for the control of the electronic blackboard corresponding to the hand shape detected by the hand shape detection unit
Gesture-based remote electronic blackboard control device comprising a.
The method of claim 1,
And an electronic blackboard controller for performing remote control on the electronic blackboard based on the generated control event.
The method of claim 1,
The camera image acquisition unit,
By invoking a calibration routine, the user checks at least one of the size and shape of the copyboard when the user looks at the copyboard at a distance from his or her location. To automatically set the corresponding scale value,
Gesture based remote copyboard control device.
The method of claim 3,
The camera image acquisition unit,
Analyzing the image related to a predetermined calibration gesture taken by the user in the course of performing the calibration routine,
When the user looks at the copyboard by calculating one or more of the diagonal length of the rectangular copyboard, the size of the angle formed when the two diagonal lines intersect, the position of each vertex, and the length of at least three of the four sides. Checking at least one of the size and shape of the copyboard of the electronic board, and setting the scale value with respect to the horizontal axis and the vertical axis to a fixed scale value or to a scale function in which the scale value varies depending on the position based on the check result;
Gesture based remote copyboard control device.
The method of claim 3,
The camera image acquisition unit,
Pan-Tilt-Zoom (PTZ) control is included so that the user's torso can be rotated or zoomed in at least one of the up, down, left and right directions with respect to the camera so that the user's upper body enters the center of the image. Execute Zoom Out,
Gesture based remote copyboard control device.
The method of claim 1,
The region of interest setting unit,
Set the region of interest based on the detected position of the hand, or set the region of movement as the region of interest using the difference image comparing the current image with the previous image acquired by the camera image acquisition unit, or skin color By using the sample to set the area where the skin color is detected as the region of interest,
Gesture based remote copyboard control device.
The method of claim 1,
The hand detector includes an individual hand detector for detecting individual predetermined hand shapes, wherein the predetermined hand shape includes at least one of a moving cursor hand shape, an executing hand shape, and a selected moving hand shape. Included-,
The predetermined individual hand gestures are detected through hierarchical steps for detecting the individual hand gestures in an order based on a preset priority.
Gesture based remote copyboard control device.
The method of claim 7, wherein
In the case of the detection of the execution hand, the hand detection unit,
Only when the execution hand gesture is subsequently detected after a predetermined preceding gesture has been detected, determining as the detection of a valid execution hand gesture,
Gesture based remote copyboard control device.
The method of claim 3,
The hand detection unit,
The position change in the image generated according to the hand shape is calculated based on the detected hand shape, and the position change in the image space is converted into the movement on the actual copyboard by reflecting the set scale value in the calculated position change amount. To be,
Gesture-based Remote Chalkboard Control Device. As a method of remotely controlling the copyboard,
By using an image of a user remote from the copyboard, at least one of the size and shape of the copyboard when the user looks at the copyboard at his / her location is checked, and the user's gesture on the viewing point of the user Setting a calibration scale value to be applied correspondingly;
Setting an area in which the user's hand is located as a region of interest using an image of a user remote from the copyboard;
Detecting a hand shape taken by a user through hierarchical image analysis of the ROI;
Generating a predetermined control event for controlling the electronic blackboard corresponding to the detected hand shape; And
Performing remote control on the electronic blackboard based on the generated control event;
Electronic blackboard remote control method comprising a.
The method of claim 10,
The setting of the calibration scale value may include:
Analyzing the image related to a predetermined calibration gesture taken by the user in the course of performing the calibration routine,
When the user looks at the copyboard by calculating one or more of the diagonal length of the rectangular copyboard, the size of the angle formed when the two diagonal lines intersect, the position of each vertex, and the length of at least three of the four sides. Checking at least one of the size and shape of the copyboard of the, and setting the scale value with respect to the horizontal axis and the vertical axis in the form of a scale function of varying the scale value according to the position or position based on the check result,
Blackboard remote control method.
The method of claim 10,
Setting the region of the user's hand as the region of interest in the acquired image,
Set the region of interest based on the detected position of the hand, or set the region of movement as the region of interest using the difference image comparing the current image with the previous image acquired by the camera image acquisition unit, or skin color By using the sample to set the area where the skin color is detected as the region of interest,
Blackboard remote control method.

Images