KR101958334B1 - Method and apparatus for recognizing motion to be considered noise - Google Patents

Abstract

In the present invention, in recognizing an operation using a sensing value through one or more sensors that detect a physical change, it is possible to reliably obtain the reliability of the operation recognition regardless of the surrounding situation by eliminating the influence of noise caused by the surrounding situation The method includes the steps of: collecting a sensing value from at least one sensor that senses a physical change; analyzing a change in the sensed value until a user operation is sensed; Extracts the feature information of the noise pattern, and performs a motion recognition process using the feature information of the predefined operation and the feature information of the noise pattern when a user operation is detected.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a method and apparatus for recognizing motion in consideration of noise,

The present invention relates to a method and an apparatus for recognizing an operation using a sensing value through at least one sensor for sensing a physical change, and more particularly, to a method and apparatus for recognizing an operation by eliminating the influence of noise caused by a surrounding situation And more particularly, to a method and apparatus for recognizing motion in consideration of noise that can ensure reliability of motion recognition at all times.

In recent years, with the introduction of an open OS, a smart phone combining a high function of a PC with a mobile phone has become popular, and various attempts have been made to utilize a high-function and high-performance smart phone.

In particular, with the development of micro-fabrication technology, advanced sensors become smaller and cheaper, and more sensors can be mounted on smart phones, and intelligent applications utilizing such sensors, such as augmented reality and 3D games, are being developed .

In addition, sensors mounted on smart phones will be able to play a key role in communicating emotions with human beings by evolving into an intelligent sensor that takes into consideration the user's physical changes and emotional states in a device that senses the surroundings. As a result, intelligent applications utilizing sensors are expected to grow even more.

There are camera (image) sensor, acoustic sensor, proximity sensor, illuminance sensor, gravity sensor, GPS sensor, acceleration sensor, gyro sensor, and geomagnetic sensor.

Among them, an acceleration sensor that detects a change of an object speed per unit time, an impact, a dynamic force such as an impact, a geomagnetic sensor that detects the azimuth angle like a compass by grasping the flow of a geomagnetic field, detects an inertial force of an object as an electric signal, And a gyro sensor for detecting the rotation angle are mainly used for motion recognition.

However, when the motion is recognized by analyzing the sensed values of the acceleration sensor, the geomagnetism sensor, and the gyro sensor, noise may be generated in the sensing value due to movement of the mobile station due to the vibration state of the terminal, bus, automobile, subway, Even if the user performs the preset predefined operation due to the noise, the user may not be able to recognize the predefined operation.

Specifically, when the vehicle is at rest, the sensed values of the acceleration sensor, the geomagnetic sensor, and the gyroscope are only slightly changed, while maintaining a substantially constant value. On the other hand, When the vehicle is in motion or in a vibrating state, the sensed values of the acceleration sensor, the geomagnetic sensor, and the gyro sensor are different from each other in the stop state due to movement occurring during movement or vibration. A component of the sensing value thus generated is referred to as noise. In a situation where such a noise occurs, even if the user takes a predetermined operation, the noise component is included in the sensing value, and the user may not recognize the operation .

Korean Patent Laid-Open No. 10-2010-0081552, published on July 19, 2010 (name: device and method for detecting motion of portable terminal)

In the present invention, in recognizing an operation using a sensing value through one or more sensors that detect a physical change, it is possible to reliably obtain the reliability of the operation recognition regardless of the surrounding situation by eliminating the influence of noise caused by the surrounding situation And to provide a motion recognition method and apparatus considering noise that can be performed.

As a solution to the above-mentioned problems, the present invention provides a sensing value collection module for collecting sensing values from at least one sensor that senses a physical change; A calculation module for calculating motion information from the collected sensing values; A noise pattern extracting module for analyzing a change in the sensing value collected from the sensing value collecting module until a user operation is detected, extracting a noise pattern, extracting feature information of the noise pattern, and generating a noise pattern management table; And an operation recognition unit that detects the user operation based on the motion information calculated from the operation module and performs motion recognition processing using the feature information of the predefined operation and the feature information of the noise pattern when a user operation is detected The present invention also provides a motion recognition device in which noise is included.

The present invention also relates to a method for detecting a physical change, comprising: collecting a sensing value from at least one sensor that senses a physical change; Extracting a noise pattern by analyzing a change in the collected sensing value until a user operation is detected; Extracting feature information of the noise pattern to generate a noise pattern management table; And performing a motion recognition process using the feature information of the predefined operation and the feature information of the noise pattern when the user operation is sensed.

According to the present invention, sensing values of a gyro sensor, an acceleration sensor, a geomagnetism sensor, and the like are analyzed to recognize a user's operation. During a standby state before a user's operation is started, a sensed value is analyzed, And by recognizing the user's action based on the extracted user's operation, it is possible to distinguish and recognize the noise and the correct user's action more accurately.

Particularly, the present invention extracts a noise pattern generated by such motion when a vibration or a vibration occurs in a user's terminal to be perceived to be motion-recognized, and performs a user's operation on the noise pattern and the pre- By comparing each of them, it is possible to distinguish noise and perform motion recognition processing, thereby reducing unnecessary arithmetic operations and resource waste, and reducing the error recognition rate, thereby improving the reliability of motion recognition results.

1 is a diagram showing a reference coordinate system and motion information used for motion recognition.
2 is a diagram illustrating a predefined operation recognized by the motion recognition apparatus according to the present invention.
FIG. 3 is a block diagram showing an operation recognition apparatus considering noise according to the present invention.
4 and 5 are flowcharts illustrating an operation recognition method in consideration of noise according to the present invention.
Figs. 6 and 7 are graphs illustrating changes in sensor measurement values in the stationary state and the moving state. Fig.
8 is an exemplary diagram of a noise pattern detected during operation recognition according to the present invention.
9 is an exemplary diagram of a noise pattern management table according to the present invention.
10 is a graph illustrating a predefined operation according to the present invention.
11 is an exemplary diagram of an operation management table according to the present invention.
12 is an exemplary diagram for explaining a noise recognition process according to the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description and the accompanying drawings, detailed description of well-known functions or constructions that may obscure the subject matter of the present invention will be omitted. It should be noted that the same constituent elements are denoted by the same reference numerals as possible throughout the drawings.

The terms and words used in the present specification and claims should not be construed to be limited to ordinary or dictionary meanings and the inventor is not limited to the concept of terminology for describing his or her invention in the best way. It should be interpreted as meaning and concept consistent with the technical idea of the present invention. Therefore, the embodiments described in the present specification and the configurations shown in the drawings are merely the most preferred embodiments of the present invention, and not all of the technical ideas of the present invention are described. Therefore, It is to be understood that equivalents and modifications are possible.

Also, terms including ordinal numbers such as first, second, etc. are used to describe various elements, and are used only for the purpose of distinguishing one element from another, Not used. For example, without departing from the scope of the present invention, the second component may be referred to as a first component, and similarly, the first component may also be referred to as a second component.

In addition, when referring to an element as being "connected" or "connected" to another element, it means that it can be connected or connected logically or physically. In other words, it is to be understood that although an element may be directly connected or connected to another element, there may be other elements in between, or indirectly connected or connected.

Also, the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. It is also to be understood that the terms such as " comprising "or" having ", as used herein, are intended to specify the presence of stated features, integers, It should be understood that the foregoing does not preclude the presence or addition of other features, numbers, steps, operations, elements, parts, or combinations thereof.

The noise recognition method and apparatus according to the present invention can be applied to various fields such as user operation recognition of a terminal, robot control, and the like. Hereinafter, a description will be made of a case where the present invention is applied to a user terminal.

The operation recognition in the user terminal is used for controlling a specific function of the user terminal by sensing a predefined specific motion (flipping, wiggling in a specific direction, drawing a specific pattern, user access) generated for the user terminal For example, the motion information of the user terminal may be expressed by a rotation angle of an azimuth angle, a pitch, and a roll with reference to a three-dimensional rectangular coordinate system as shown in FIG.

1 shows a reference coordinate system and motion information used for motion recognition. When the horizontal direction of the user terminal is an X axis, the vertical direction is a Y axis, and the width direction is a Z axis, the azimuth is a direction 0 to 360 ° or -180 to 180 ° with respect to the Z axis along the horizontal axis, the vertical axis, the vertical axis, the vertical axis, the vertical axis, The roll is a vertical axis rotation angle, which is expressed as -180 to 180 degrees depending on the form in which the user terminal is laid out.

Of course, in the motion recognition method and apparatus according to the present invention, the reference coordinate system and the method of expressing the motion can be changed, and the above definition is only an example.

In the motion recognition method and apparatus according to the present invention, the above-described motion information (rotation angles for x, y, and z axes) is calculated using sensing values measured by one or more sensors that detect physical changes, The motion information is compared with the motion recognition condition for the predefined motion, and motion recognition is performed.

The predefined operations recognized by the method and apparatus for recognizing an operation according to the present invention can be a combination of one or more unit operations. Here, the unit operation means an operation in which the predefined operation is subdivided based on at least one of direction, time, and speed.

As shown in FIG. 2, when an operation of moving a terminal is defined, it can be divided into a unit operation 1, a unit operation 2, and a unit operation 3 according to a direction. Therefore, when the unit operation 1, the unit operation 2, and the unit operation 3 are sequentially recognized, it can be determined that the operation is recognized. Therefore, the motion recognition condition for the motion recognition can be set as a unit motion unit.

An apparatus and method for recognizing motion in consideration of noise according to the present invention will be described based on the above description.

FIG. 3 is a block diagram showing an operation recognition apparatus considering noise according to the present invention.

Referring to FIG. 3, the motion recognition apparatus 100 according to the present invention collects sensing values output from one or more sensors 10 that detect a physical change, compares and analyzes the sensed values, Extracts a corresponding noise pattern, and recognizes the predefined operation in consideration of the extracted noise pattern. At this time, the sensing value may be collected at a predetermined sampling period.

The motion recognition apparatus 100 may include a sensing value collection module 110, a calculation module 120, a noise pattern extraction module 130, and a motion recognition module 1400.

The sensing value collecting module 110 collects sensed values output from the at least one sensor 10 at regular intervals. At this time, the sensing value output from the at least one sensor 10 has a different unit value depending on the sensor. For example, the sensing value of the illuminance sensor indicates the amount of illumination (Lux), the sensing value of the proximity sensor indicates the distance to and / or proximity to a nearby object, and the sensing value of the 3-axis acceleration sensor is 3 Axis, the sensing value of the three-axis gyro sensor indicates the angular velocity in the direction of the three axes (X, Y, Z), and the sensing value of the geomagnetism sensor indicates the acceleration value in the direction of the axis Direction. The gravitational acceleration can be measured based on the sensing value of the acceleration sensor, thereby determining how the user terminal is located. For example, when the terminal is placed on the ground, the z-axis is affected by the gravitational acceleration, and the z-axis sensing value of the acceleration sensor is about 1G (= 9.8 m / s ² ). The gyro sensor measures the rotational angular velocity of each of three axes, through which the user can ascertain the direction in which direction the user terminal rotates.

The calculation module 120 combines one or more sensing values collected by the sensing value collection module 110 to calculate motion information. More specifically, the calculation module 120 converts the sensed values collected by the rotation vector and rotation matrix calculations into rotational angles with respect to the x, y, and z axes of a predetermined coordinate system as shown in FIG.

The motion recognition apparatus 100 according to the present invention periodically collects sensing values of at least one sensor 10 through the sensing value collection module 110 and the calculation module 120 in a standby state for recognizing an operation, The motion recognition module 140 calculates motion information (rotation angle) based on the motion information calculated by the calculation module 120, and determines whether or not user movement has occurred, that is, .

Meanwhile, the noise pattern extracting module 130 extracts a noise characteristic according to the state of the terminal to which the motion recognition apparatus 100 is currently applied. Specifically, the noise pattern extracting module 130 continuously monitors one or more sensed values collected by the sensed-value collecting module 110 so as to extract the sensed values from a time point at which one or more sensors 10 are enabled The noise pattern is extracted from the section up to the time when the definition operation is generated. Here, the section up to the start of the user operation is referred to as a UE status check interval. The noise pattern is extracted as a minimum unit of movement repeated in the same form in the UE status confirmation interval, for each sensing value type.

Lastly, the motion recognition module 140 is configured to recognize a predefined motion using the motion information output from the calculation module 200. Specifically, in the present invention, the motion recognition module 140 compares motion information calculated from the calculation module 120 with a predefined motion to determine whether a user motion has occurred, and determines that a user motion has occurred The noise pattern extraction module 130 removes the noise pattern from the sensing value corresponding to the user operation and then compares the noise pattern with the predefined operation to determine whether the user operation is a predefined operation.

The operation recognition process considering the noise pattern by the motion recognition apparatus 100 of the present invention configured as described above will be described with reference to FIG. 4 to FIG. 4 and 5 are flowcharts illustrating a method of recognizing an operation in consideration of noise according to the present invention, and FIGS. 6 to 12 are illustrations for explaining the operation.

4, the motion recognition apparatus 100 according to the present invention periodically receives sensing values of at least one sensor 10 through the sensing value acquisition module 110 in the operation recognition standby state (S105) (S110).

One or more sensing values collected by the sensing value acquisition module 110 may be, for example, x, y, z axis sensing values of a three-axis gyro sensor, x, y, z axis sensing values of a three- . The at least one sensed value is input to a computation module 120 that computes motion information (e.g., a rotation angle) from the sensed value or values, and then the motion recognition module 140 It is determined whether a priority user action is generated based on the motion information (S115).

In addition, the sensing value collected by the sensing value collection module 110 is input to the noise pattern extraction module 130. The noise pattern extraction module 130 extracts the sensing value from the sensing value acquisition module 110 before the user operation starts The change pattern of the sensing value generated according to the surrounding situation is checked (S120), and the noise pattern is extracted (S125).

For example, FIG. 6 is a graph showing a sensing value generated by the at least one sensor 10 when a user terminal equipped with one or more sensors 10 is in a stopped state, FIG. And a sensing value generated in the at least one sensor 10 when the sensor 10 is in a sensing state. In particular, Figures 6 and 7 illustrate the sensing values of the acceleration sensor and the gyro sensor.

When the user terminal is in a stationary state, the sensing values of the x, y, and z axes generated by the acceleration sensor and the gyro sensor appear almost linearly without significant change as shown in FIG. 6, , And when it is moving, as shown in Fig. 7, a certain change pattern appears repeatedly. Accordingly, the state of the user terminal can be grasped by checking the change state of the sensing value collected by the at least one sensor 10. [

In step S120, it is determined whether or not a noise occurs due to movement or vibration by checking a change pattern of the sensing value.

In step S125, the noise pattern extraction module 130 extracts a noise pattern from a change pattern appearing in the sensing value due to movement caused by movement or vibration of the user terminal. In step S125, As shown in FIG. 7, when a predetermined change is shown in the collected sensing value, a noise pattern which is the minimum unit of the repeated change in the same form is extracted. The noise pattern may be extracted for each sensing value. For example, the x, y and z axis sensing values of the acceleration sensor, the x, y and z axis sensing values of the gyro sensor, and the noise pattern for each are extracted.

The noise pattern extracting module 130 extracts feature information of the extracted noise pattern to generate a noise pattern management table (S130). The characteristic information of the noise pattern may include at least one measurement point at which the increase / decrease type (+/-) is switched and a mean rotation angle at each measurement point in the noise pattern when the change in the sensing value is out of a certain range have. In addition, the feature information of the noise pattern may further include information indicating a degree of change of the sensing value. The information indicating the degree of change may be represented by, for example, a standard deviation of an average rotation angle of at least one measurement time point. The characteristic information of the noise pattern described above is recorded in the noise pattern management table.

FIG. 8 is an example of a noise pattern extracted by the noise pattern extracting module 130 according to the present invention, and is a noise pattern extracted from the z-axis sensing value of the acceleration sensor shown in FIG.

P _N0 (Start), P _N1 , P _N2 , P _N3 , P _N4 , and P _N5 (End) as measurement points at which a change over a certain range occurs in the noise pattern shown in FIG. Can be extracted. In addition, the noise pattern extracting module 130 grasps the number of accumulated noise patterns during the terminal state checking period, which is the period before the user operation starts in the operation recognition waiting state, (Rotation angle) of the plurality of noise patterns appearing during the terminal status confirmation period for each measurement time is calculated, and the average motion information (rotation angle) , And generates a noise pattern management table. In the noise pattern management table, time information [ms] (O ms) and average rotational angle [degree] (□) are mapped and stored for each measurement time point. In addition, the noise pattern extracting module 130 obtains a standard deviation (DELTA deg) of the average motion information calculated at one or more measurement times and records the obtained standard deviation in the noise pattern management table of FIG.

The noise pattern management table as shown in FIG. 9 can be generated corresponding to one or more sensing values.

If it is determined in step S115 that the user operation is to be performed, the sensing value collected by the sensing value collection module 110 is transmitted to the motion recognition module 140 so as to reduce unnecessary processing. And performs a filtering process.

More specifically, the operation recognition module 140 basically includes an operation management table in which the feature information of the predefined operation is recorded for the predefined operation to be recognized. 10 and 11 are diagrams illustrating a predefined operation and a corresponding operation management table. The operation management table also shows that the degree of change extracted from the change pattern of one or more reference sensing values corresponding to the predefined operation is within a certain range At least one measuring point at which the increasing / decreasing type (+/-) is switched off, time information about the at least one measuring point, and motion information (rotational angle) are mapped and set.

In this state, the motion recognition module 140 compares the sensed values collected from the start of the user operation from the sensing value acquisition module 110 with the feature information of the predefined motion of the operation management table, y, and z axes coincide with the predefined motion (S140). That is, it is determined whether the moving direction from the operation start point to the operation completion point coincides with the predefined operation. If it is determined that the direction of the user operation does not coincide with any one of the x, y, and z axes, the motion recognition process is not performed based on the determined noise.

When the direction of at least one of the x, y, and z axes of the user operation coincides with the predefined operation, the motion recognition module 150 refers to the noise pattern management table, That is, the motion information of the user motion detected based on the axis having the smallest standard deviation is compared with the motion information stored in the noise pattern management table, and the difference is calculated to determine whether the difference is within a predetermined range (S145, S150). If it is determined that the difference between the motion information of the user's operation and the motion information defined in the noise pattern management table is within a certain range, the motion recognition process is discontinued. According to this, it is possible to reduce the erroneous recognition and reduce unnecessary arithmetic processing and resource waste.

If it is determined that the difference between the motion information of the user's operation and the motion information defined in the noise pattern management table is out of a predetermined range as a result of the determination, it is determined that the noise is not noise, (S155). In step S155, the motion recognition process may be performed by a method of removing a noise component and comparing the change pattern with a change pattern of the predefined operation.

The above-described step S155 may be performed as shown in FIG.

Referring to FIG. 5, in step S155, the motion recognition module 140 first determines whether or not the noise pattern management table generated in the noise pattern extraction module 130 and the noise pattern management table The sensed values are extracted, and the extracted sensed values are connected to extract a change pattern of the user's operation (S205).

Assuming, for example, that as shown in FIG. 12, the noise pattern 12 is extracted and the predefined operation 13 is set and the sensing value is collected as shown in FIG. 14 corresponding to the user operation, The motion recognition module 140 is configured to recognize six measurement points P _N0 through P _N5 extracted from the noise pattern 12 and seven measurement points P ₀ through P ₆ extracted from the pre- The sensed value of the user's operation is extracted. In the lower graph of FIG. 12, the red dot represents the sensing value of the user operation extracted at the measurement time points, P _N0 to P _N5 , and P ₀ to P ₆ , respectively. Then, the extracted sensing values are connected in chronological order to extract the user operation pattern 15. At this time, the measurement time at which the sensing value is suddenly changed for a predetermined time is excluded. In Fig. 12, the points corresponding to the measurement points P ₁ and P _N1 are excluded.

Then, the operation recognition module 140 compares the extracted user operation pattern 15 with the change pattern of the predefined operation 13 to determine whether it matches the predefined operation.

To this end, the motion recognition module 140 extracts the slope and the increase / decrease form between the measurement points in the user operation pattern 15 (S210).

The matching rate is calculated by comparing the number of increase / decrease conversion of the extracted user operation pattern 15 with the number of increase / decrease conversion in the dictionary definition operation 13 (S215). The calculated matching rate is compared with the reference value. It is recognized as a predefined operation (S225). The reference value may be set as shown in Equation (1) below.

By recognizing the user's operation through the above-described process, it is possible to efficiently recognize the predefined operation even in a situation where noise occurs due to movement or vibration, thereby minimizing false recognition due to noise.

The motion recognition method capable of adjusting the sensitivity according to the present invention can be implemented in a form of software readable by various computer means and recorded in a computer-readable recording medium. Here, the recording medium may include program commands, data files, data structures, and the like, alone or in combination. Program instructions to be recorded on a recording medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. For example, the recording medium may be an optical recording medium such as a magnetic medium such as a hard disk, a floppy disk and a magnetic tape, a compact disk read only memory (CD-ROM), a digital video disk (DVD) Includes a hardware device that is specially configured to store and execute program instructions such as a magneto-optical medium such as a floppy disk and a ROM, a random access memory (RAM), a flash memory, do. Examples of program instructions may include machine language code such as those generated by a compiler, as well as high-level language code that may be executed by a computer using an interpreter or the like. Such a hardware device may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

Further, the functional operations and subject matter implementations described herein may be implemented in other types of digital electronic circuitry, or may be implemented in computer software, firmware, or hardware, including the structures disclosed herein and their structural equivalents, ≪ / RTI > Implementations of the subject matter described herein may be embodied in one or more computer program products, i. E. One for computer program instructions encoded on a program storage medium of the type for < RTI ID = 0.0 & And can be implemented as a module as described above. The computer-readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composition of matter that affects the machine readable propagation type signal, or a combination of one or more of the foregoing.

In addition, while the specification contains a number of specific implementation details, it should be understood that they are not to be construed as limitations on the scope of any invention or claim, but rather on the features that may be specific to a particular embodiment of a particular invention Should be understood as an explanation. Certain features described herein in the context of separate embodiments may be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented in multiple embodiments, either individually or in any suitable subcombination. Further, although the features may operate in a particular combination and may be initially described as so claimed, one or more features from the claimed combination may in some cases be excluded from the combination, Or a variant of a subcombination.

Likewise, although the operations are depicted in the drawings in a particular order, it should be understood that such operations must be performed in that particular order or sequential order shown to achieve the desired result, or that all illustrated operations should be performed. In certain cases, multitasking and parallel processing may be advantageous. Also, the separation of the various system components of the above-described embodiments should not be understood as requiring such separation in all embodiments, and the described program components and systems will generally be integrated together into a single software product or packaged into multiple software products It should be understood.

Certain embodiments of the subject matter described herein have been described. Other embodiments are within the scope of the following claims. For example, the operations recited in the claims may be performed in a different order and still achieve desirable results. By way of example, the process illustrated in the accompanying drawings does not necessarily require that particular illustrated or sequential order to obtain the desired results. In certain implementations, multitasking and parallel processing may be advantageous.

The description sets forth the best mode of the invention, and is provided to illustrate the invention and to enable those skilled in the art to make and use the invention. The written description is not intended to limit the invention to the specific terminology presented. Thus, while the present invention has been described in detail with reference to the above examples, those skilled in the art will be able to make adaptations, modifications, and variations on these examples without departing from the scope of the present invention.

Therefore, the scope of the present invention should not be limited by the described embodiments but should be defined by the claims.

According to the present invention, sensing values of a gyro sensor, an acceleration sensor, a geomagnetism sensor, and the like are analyzed to recognize a user's operation. During a standby state before a user's operation is started, a sensed value is analyzed, And by recognizing the user's action based on the extracted user's operation, it is possible to distinguish and recognize the noise and the correct user's action more accurately.

Particularly, the present invention extracts a noise pattern generated by such motion when a vibration or a vibration occurs in a user's terminal to be perceived to be motion-recognized, and performs a user's operation on the noise pattern and the pre- By comparing each of them, it is possible to distinguish noise and perform motion recognition processing, thereby reducing unnecessary arithmetic operations and resource waste, and reducing the error recognition rate, thereby improving the reliability of motion recognition results.

10: Sensor
100: motion recognition device
110: Sensing value acquisition module
120: Operation module
130: Noise pattern extraction module
140: motion recognition module

Claims (11)

A sensing value collection module for collecting sensing values from at least one sensor that senses a physical change;
A calculation module for calculating motion information from the collected sensing values;
A noise pattern extracting module for analyzing a change in the sensing value collected from the sensing value collecting module until a user operation is detected, extracting a noise pattern, extracting feature information of the noise pattern, and generating a noise pattern management table; And
An operation recognition module that detects the user operation based on the motion information calculated from the operation module and performs motion recognition processing using feature information of the predefined operation and feature information of the noise pattern when a user operation is detected, Wherein the motion recognition device includes: The method according to claim 1,
Characterized in that the feature information of the noise pattern includes at least one measurement point at which the variation of the sensing value is out of a predetermined range and in which the increase / decrease type is switched, average motion information at the at least one measurement point, Wherein the motion recognition device includes at least one of a motion detection device and a motion detection device. The apparatus of claim 1, wherein the motion recognition module
Comparing the sensed value corresponding to the user action with the predefined action to determine whether the direction of at least one axis among the x, y, and z axes coincide with each other and the axis having the smallest standard deviation in the noise pattern management table as a reference The difference between the feature information of the noise pattern and the feature information of the user's operation is found and the operation recognizing process is performed when the difference is out of a certain range. 3. The apparatus of claim 2, wherein the motion recognition module
Extracting a sensed value of a user operation measured at a measuring point extracted from the noise pattern and a measuring point extracted at the predefined operation, connecting sensed values of the extracted user operation in chronological order, And comparing the extracted user operation pattern with a change pattern of the predefined operation to determine whether the user operation is a predefined operation. 5. The apparatus of claim 4, wherein the motion recognition module
Wherein when the sensed values of the extracted user actions are connected in chronological order, a sensed value whose value difference is changed by more than a reference value is removed. 6. The apparatus of claim 5, wherein the motion recognition module
Comparing the number of times of conversion of the slope between each sensing value of the user operation pattern with the number of times of conversion of the slope code of the variation pattern of the predefined operation and judging that the user operation is a predefined operation when the matching rate is not less than the reference value The motion recognition device taking noise into consideration. 7. The method of claim 6,

Wherein the motion recognition unit recognizes the noise. Collecting a sensing value from at least one sensor that senses a physical change;
Extracting a noise pattern by analyzing a change in the collected sensing value until a user operation is detected;
Extracting feature information of the noise pattern to generate a noise pattern management table;
And performing a motion recognition process using the feature information of the predefined motion and the feature information of the noise pattern when the user motion is detected. 9. The method according to claim 8,
At least one measurement point at which the variation of the sensed value is out of a predetermined range and in which the increase / decrease mode is switched, average motion information at the at least one measurement point, and standard deviation of the average motion information at the one or more measurement points A motion recognition method in which noise is considered. 9. The method according to claim 8,
Comparing the sensing value corresponding to the user operation with the predefined operation to determine whether the directions of at least one of the x, y, and z axes coincide with each other; And
Further comprising the step of determining a difference between the feature information of the noise pattern and the feature information of the user operation based on the axis having the smallest standard deviation in the noise pattern management table and determining whether the difference is out of a predetermined range,
Wherein the motion recognizing process is performed when the directions of the at least one axis coincide with each other and the difference between the feature information of the noise pattern and the feature information of the user operation deviates from a certain range. 10. The method according to claim 9,
Extracting a sensed value of a user operation measured at a measuring point extracted from the noise pattern and at a measuring point extracted in the predefined operation;
Extracting the user action pattern by connecting sensed values of the extracted user actions in chronological order;
Comparing the extracted user action pattern with a change pattern of the predefined action, and determining whether the user action is a predefined action.

Images