KR101503017B1 - Motion detection method and apparatus - Google Patents

Abstract

A method and apparatus for detecting non-contact motion using an image sensor is disclosed. According to an embodiment of the present invention, there is provided a motion detection method including: receiving a sequential frame image of a subject from an image sensor; Generating a difference image between a frame image of the input current frame and a frame image of a previous frame; Projecting the difference image onto a reference axis to obtain projection data; Extracting feature information from the projection data; Analyzing a trajectory of the feature information; And acquiring motion information of the subject using the analyzed trajectory. According to this, non-contact motion information of the subject can be obtained by using the image sensor.

Image sensor, motion, difference image, projection, trajectory

Description

[0001] The present invention relates to a motion detecting method and apparatus,

The present invention relates to a motion detection method and apparatus, and more particularly, to a method and apparatus for detecting non-contact motion using an image sensor.

2. Description of the Related Art Portable electronic apparatuses equipped with an imaging unit (e.g., a camera module) and a sound source reproducing unit (e.g., an MP3 file reproducing module) are provided according to recent breakthroughs in electronics and communication engineering. Accordingly, a user can enjoy various functions such as internet access, video communication, video message output, digital music listening, and satellite broadcast viewing using portable electronic devices having various functions. In order to enjoy these various functions, efficient key input for selecting various functions is required.

However, in the case of a conventional portable electronic device, a user can input various key inputs (for example, a call button and a wireless Internet access button in a mobile communication terminal, a file play button and a file selection button in an MP3 player, A volume control button, or the like), or inputting a key through a screen touch.

In addition, in order to detect movement of a finger or the like in a non-contact manner using an imaging unit provided in a portable electronic device, it is necessary to receive reflected light due to finger movement using an independent light source such as an infrared light source.

Accordingly, the present invention provides a motion detection method and apparatus capable of acquiring non-contact motion information of a subject using an image sensor.

In addition, the present invention provides a motion detection method and apparatus capable of acquiring motion information and generating an instruction using the sequential frame information without imposing an operational burden on the processor.

According to an aspect of the present invention, there is provided a motion detection method for detecting a motion of a subject and a recording medium on which a program for performing the motion detection is recorded.

According to an embodiment of the present invention, there is provided a motion detection method including: receiving a sequential frame image of a subject from an image sensor; Generating a difference image between a frame image of the input current frame and a frame image of a previous frame; Projecting the difference image onto a reference axis to obtain projection data; Extracting feature information from the projection data; Analyzing a trajectory of the feature information; And acquiring motion information of the subject using the analyzed trajectory.

The step of generating the difference image may generate the difference image using a difference between pixel values of pixels corresponding to each other between a frame image of the current frame and a frame image of the immediately preceding frame as pixel values.

The obtaining of the projection data may obtain the projection data by summing the pixel values of the pixels in the difference image toward the reference axis. Or the step of acquiring the projection data may acquire, by the projection data, the number of pixels in the difference image whose pixel value is greater than or equal to a threshold value toward the reference axis.

Wherein the extracting of the feature information comprises: grouping the projection data around a detection reference projection point; And generating the feature information using the difference of the sum of the projection data within each group divided centering on the detection reference projection point.

The step of analyzing the locus may determine whether or not the motion of the subject is determined by determining whether the accumulated feature information has a predetermined shape over time. Here, the predetermined type may be a sinusoidal waveform.

The reference axis may include a horizontal axis and a vertical axis of the difference image. In the step of analyzing the locus, when it is determined that there is motion of the subject, the step of analyzing the motion may determine an axis having the dominant trajectory of the horizontal axis and the vertical axis as the main direction of the motion of the subject .

According to another aspect of the present invention, there is provided a motion detection apparatus for detecting a motion of a subject.

According to an embodiment of the present invention, there is provided a motion detection apparatus including: an image input unit that receives a sequential frame image of a subject from an image sensor; A difference image generation unit for generating a difference image between a frame image of a current frame and a frame image of a previous frame inputted by the image input unit; An image projection unit for projecting the difference image onto a reference axis to obtain projection data; A feature information extracting unit that extracts feature information from the projection data; a locus analyzing unit that analyzes a locus of the feature information; And a motion analyzer for obtaining motion information of the subject using the analyzed locus.

The difference image generating unit may generate the difference image in which a difference between pixel values of pixels corresponding to each other between a frame image of the current frame and a frame image of the immediately preceding frame is a pixel value.

The image projection unit may obtain the projection data by summing pixel values of pixels in the difference image toward the reference axis. Alternatively, the image projection unit may obtain, as the projection data, the number of pixels in the difference image whose pixel value is greater than or equal to a threshold value toward the reference axis.

The feature information extracting unit may group the projection data around a detection reference projection point and generate the feature information by using a difference between sums of projection data within each group divided based on the detection reference projection point.

The trajectory analysis unit may determine whether or not the subject has motion by determining whether the accumulated feature information has a predetermined shape over time. Here, the predetermined type may be a sinusoidal waveform.

The reference axis may include a horizontal axis and a vertical axis of the difference image. When it is determined by the trajectory analysis unit that there is motion of the subject, the motion analyzer can determine an axis having a dominant trajectory of the horizontal axis and the vertical axis as the main direction of the motion of the subject.

The motion detection apparatus may further include an automatic exposure unit that adjusts an exposure value of the image sensor according to a surrounding environment to input the frame image at a constant frame rate.

Other aspects, features, and advantages will become apparent from the following drawings, claims, and detailed description of the invention.

The motion detection method and apparatus according to the present invention can acquire non-contact motion information of a subject using an image sensor.

In addition, it is possible to acquire motion information and generate a command using the motion information without imposing a computational burden on the processor by using the sequential frame information.

BRIEF DESCRIPTION OF THE DRAWINGS The present invention is capable of various modifications and various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. It is to be understood, however, that the invention is not to be limited to the specific 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.

The terms first, second, etc. may be used to describe various components, but the components should not be limited by the terms. The terms are used only for the purpose of distinguishing one component from another.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises" or "having" and the like are used to specify that there is a feature, a number, a step, an operation, an element, a component or a combination thereof described in the specification, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram of a motion detection apparatus according to an embodiment of the present invention. Referring to FIG. 1, an image sensor 10, an automatic exposure unit 15, and a motion detection device 100 are shown.

The image sensor 10 converts an optical image (i.e., an optical subject signal) input through an external lens into an electrical signal and outputs the electrical signal. Here, an electrical signal output every frame is referred to as a frame image. The image sensor 10 outputs a frame image having a constant frame rate or a varying frame rate. Hereinafter, for ease of understanding and explanation of the present invention, it is assumed that a frame image has a constant frame rate. The image sensor 10 may be a CCD (Charge Coupled Device) type or a CMOS (Complementary Metal-Oxide-Silicon) type.

The motion detection apparatus 100 includes an image input unit 110, an image storage unit 115, a difference image generation unit 120, an image projection unit 130, a feature information extraction unit 140, a feature information storage unit 145, A locus analyzing unit 150, and a motion analyzing unit 160.

The image input unit 110 receives a sequential frame image from the image sensor 10. The received frame image may be a color image or a monochrome image. In the case of a color image, the color image can be used as it is or converted to a monochrome image. In the case of a black-and-white image, there is an effect that the overall calculation amount of the motion detection device 100 can be reduced in detecting the motion of the subject.

The received frame image is stored in the image storage unit 115. Assuming that the current frame is the n-th frame, a frame image corresponding to the frame (n-1) th frame preceding the frame (n-th frame) corresponding to the currently input frame image is stored in the image storage unit 115 have.

The difference image generation unit 120 generates a difference image using the difference between the frame image of the nth frame (nth frame image) of the current frame input to the image input unit 110 and the nth frame of the previous frame read from the image storage unit 115 (N-1) < th > frame image) is used to generate a difference image. For any pixel position, the difference between the pixel values at the corresponding pixel positions in the two frame images is taken as the pixel value at that pixel position in the difference image. The pixel value of the corresponding pixel in the difference image is preferably the absolute value of the difference between the pixel values of the corresponding pixel in the two frame images as shown in Equation 1 below.

Here, F _diff is the pixel value in the difference image, F (n) is the pixel value in the nth frame image, and F (n-1) is the pixel value in the n-1th frame image.

When generating the difference image, operations such as cropping and masking of the frame image may be additionally preceded according to the motion detection width. Here, the motion detection width means a motion direction, a motion size, and the like of a subject that can be detected by the motion detection apparatus 100 according to the present invention.

In one embodiment, when it is desired to detect not only the left-right direction or the up-down direction but also the 360-degree whole direction, the masking can be performed using the circular mask around the detection reference point. In another embodiment, high weighting may be given to pixels located close to the detection reference point, and low weighting may be given to pixels located at a distance.

In the embodiment of the present invention, when detecting the motion of a subject, it is assumed that there is motion when the subject moves through a specific point or line. In this case, a specific point or line that is the center is referred to as a detection reference point. The detection reference point may be the center point of the image sensor 10 or the center point of the image captured by the image sensor 10. [ Further, the detection reference point may be a pixel on the image designated by the user. In addition, the detection reference point may be one or more.

In addition, the difference image generation unit 120 may perform adaptive thresholding to improve the contrast of the entire difference image after generating the difference image.

The image projection unit 130 projects a difference image generated by the difference image generation unit 120 onto a reference axis. The difference image is projected on the reference axis, so that the projection data on the reference axis is obtained. Here, the reference axis may be any axis within the difference image, and may be the horizontal axis or the vertical axis of the difference image. The reference axis may be one or two or more. Hereinafter, it is assumed that the reference axis is the horizontal axis (X axis) and the vertical axis (Y axis) of the difference image, but the present invention is not limited thereto.

Assuming the X axis and Y axis perpendicular to each other as the reference axis, the pixels of the difference image can be represented by (x, y).

In one embodiment, the projection data (Y _proj (y)) on the Y-axis with arbitrary y-coordinates is the _sum of all the pixels (x ₁ , y), (x ₂ , y) (x ₃ , y), ..., (x _k1 , y). The projection data (X _proj (x)) on the X-axis having an arbitrary x-coordinate is obtained by interpolating all the pixels (x, y ₁ ), (x, y ₂ ) ₃ ), ..., (x, y _k2 )). This can be expressed by the following equation (2).

In another embodiment, the projection data (Y _proj (y)) on the Y-axis with arbitrary y-coordinates is obtained by interpolating all pixels (x ₁ , y), (x ₂ , y) (x ₃ , y), ..., (x _k1 , y). Then, the projection data (X _proj (x)) on the X-axis having the arbitrary x-coordinate is the sum of all the pixels ((x, y ₁ ), (x, y ₂ ) ₃ ), ..., (x, y _k2 )).

In one embodiment, the projection data on the X-axis allows information on motion in the X-axis direction to be obtained, and the projection data on the Y-axis can acquire information on the motion in the Y-axis direction . It is a matter of course that the direction is determined in order to determine a criterion for extracting feature information later, and conversely, the opposite case is of course possible.

The feature information extraction unit 140 groups the projection data acquired by the image projection unit 130 according to a predetermined method. In one embodiment, the detection reference points projected on the X-axis and / or Y-axis can be grouped into two groups on both sides of a detection reference point. Then, feature information is generated through the difference of each group.

In one embodiment, there is one detection reference point and two detection reference points are grouped into two groups around the detection reference point. In another embodiment, three detection reference points are grouped based on two or more detection reference points projected on two or more detection reference points, Grouping can be performed. In this case, the feature information can be generated through the difference of each group. For example, the feature information may be generated by dividing the group into the highest three groups and the remaining groups, or the feature information may be generated using the projection data of the uppermost group and the adjacent group, The feature information can be generated using the projection data of the group.

Hereinafter, when there is one detection reference point, that is, when the detection reference projection point is one, it is assumed that the feature information is generated by being grouped into two groups.

The generated feature information is stored in the feature information storage unit 145 and used by the trajectory analysis unit 150 to acquire motion information such as a trajectory. The feature information extracting unit 140 generates feature information by grouping the projection data on the X axis and the Y axis, respectively.

The locus analyzing unit 150 determines whether or not the motion of the subject is determined by determining whether or not the trajectory of the feature information with a lapse of time has a predetermined shape (for example, a sine waveform) based on the feature information. When the size is equal to or greater than a predetermined threshold value, it is determined that there is motion in a specific direction centering on the detection reference point.

When the preset shape is a sinusoidal waveform, there is an advantage that the trajectory of the feature information can be grasped by a relatively simple calculation. The trajectory analyzer 150 analyzes the trajectory of the feature information with respect to the reference axis and analyzes the trajectory of the feature information with respect to each reference axis when there are two or more reference axes (i.e., the X axis and the Y axis are reference axes) do.

When it is determined by the locus analyzing unit 150 that there is motion of the subject, the motion analyzer 160 grasps the characteristics of the locus and analyzes the motion direction of the subject on the reference axis. The feature information is used to identify the area where the change is large for each frame among the difference images. The trajectory of the feature information reflects the sign change of the difference value according to the change of the corresponding region. In other words, when the trajectory changes from a negative value to a positive value and from a positive value to a negative value, even if a similar sinusoidal waveform is used, Different.

Also, the motion analyzer 160 analyzes the results of the X-axis and the Y-axis analyzed by the trajectory analyzer 150 when the reference axis is two or more, and selects the one having a relatively dominant shape. If the result of the X-axis is dominant, it is determined that the motion of the subject is in the X-axis direction. If the result of the Y-axis is predominant, the motion of the subject is determined to be in the Y-axis direction. That is, the motion analyzer 160 determines the two-dimensional directionality of the motion of the subject.

The X-axis and the Y-axis have a similar shape, and if the difference is within a predetermined range, it can be determined that the subject moves with an oblique line to the X-axis and the Y-axis.

The motion analyzing unit 160 outputs the motion information (including the motion direction) of the subject analyzed, and the processor connected to the rear end uses the motion information. For example, by using the motion information of the subject, it is possible to display the same effect as moving the mouse cursor in the left-right direction or the up-down direction, or to select any one of a plurality of selection objects arranged in the left-right direction or the up-down direction .

According to another embodiment of the present invention, the automatic exposure unit 15 may be additionally connected to the image sensor 10. In the above description, it is assumed that the frame rate is constant when the motion detection apparatus 100 acquires motion information through the frame image. The automatic exposure unit 15 adjusts the exposure value so that the image sensor 10 can output the frame image at a constant frame rate.

In the case of an image sensing apparatus such as a digital camera having an image sensor 10, a sufficient amount of light is input to the image sensor 10 by adjusting the frame rate and the exposure value according to the surrounding environment (for example, brightness). For example, assuming that the exposure value is fixed, it is 40 fps (Frame Per Second) when the surroundings are bright as in the daytime, but may be 5 to 10 fps when the surroundings are dark like nighttime. In this case, the frame rate changes depending on the surrounding environment, and it may be difficult for the motion detection apparatus 100 requiring a constant frame rate to obtain accurate motion information. Accordingly, the automatic exposure unit 15 may maintain the frame rate constant while varying the exposure value according to the surrounding environment.

1, the automatic exposure unit 15 is connected to the image sensor 10 separately from the motion detection device 100. However, the automatic exposure unit 15 may be provided in the motion detection device 100 .

In an embodiment of the present invention, the motion detection apparatus 100 photographs a subject using the image sensor 10 under visible light conditions and analyzes the motion of the subject by frame, without having an independent light source such as an infrared light source, .

In another embodiment, the motion detection apparatus 100 may have an auxiliary light source to acquire a sufficient amount of light when the motion detection apparatus 100 has the automatic exposure unit 15 and can not acquire a sufficient amount of light under the visible light condition.

Hereinafter, a method of detecting the motion of a subject in the motion detection apparatus 100 will be described in more detail with reference to FIG. 2 and subsequent drawings. For convenience of understanding and explanation of the invention, it is assumed that the motion detection method is performed by the motion detection apparatus 100. [

FIG. 2 is a flowchart of a motion detection method according to an embodiment of the present invention, FIG. 3 is a diagram illustrating a method of generating a difference image, FIG. 4 is an exemplary view of a difference image, FIG. 6 is a diagram showing a method of extracting feature information, and FIG. 7 is a diagram showing a waveform of a feature information set.

Here, it is assumed that the reference axis is the horizontal axis (X axis) and the vertical axis (Y axis) of the difference image, and the detection reference point is the center point of the difference image.

The motion detection apparatus 100 receives a frame image from the image sensor 10 (step 200). The received frame image may be a color image or a monochrome image. If it is a color image, it can switch to a black and white image. The conversion of the color image into the black and white image is obvious to those skilled in the art, and a detailed description thereof will be omitted.

A difference image is generated between the frame image of the current frame (nth frame image) and the frame image of the previous frame (n-1th frame image) (step 210). the difference between the pixel values of the pixels of the n-th frame image and the corresponding pixels of the (n-1) -th frame image is set as the pixel value of the corresponding pixel of the difference image.

Referring to Figure 3, n-1 th frame image Frame (n-1) and the n th frame image Frame (n) n-1-th difference image by performing a subtraction operation with respect to each pixel in F _diff (n-1 ). Each pixel of the difference image has a pixel value through Equation (1).

For the difference image 400, a certain area is grouped according to the motion detection width around the detection reference point 410 (refer to the Grouping area in FIG. 4), a weight is given to a certain area in the group (refer to the weighting area in FIG. 4) You may.

For the generated difference image, projection data obtained by projecting each pixel on two axes perpendicular to each other is obtained (step 220). When the horizontal axis of the difference image is referred to as X axis and the vertical axis as Y axis, two mutually perpendicular axes can be X axis and Y axis. That is, it is possible to obtain the projection data 510x on the X-axis and the projection data 510y on the Y-axis (see FIG. 5).

The projection data may be the sum of pixel values as in Equation (2), or may be the number of pixels having a certain value or more.

Based on the projection data, feature information in the X-axis direction and feature information in the Y-axis direction are extracted (step 230).

A method of acquiring feature information in the X-axis direction will be described with reference to FIG. 6, and the feature information in the Y-axis direction can also be obtained by the same method.

The projection data (X _proj ) in the X-axis direction is grouped on both sides around the detection reference point. The sum of the projection data belonging to the A group is X _{projsum_a} , and the sum of the projection data belonging to the B group is X _{projsum_b} . Here, it is assumed that the detection reference point is the center point of the difference image, and the number of the projection data belonging to the A group and the number of the projection data belonging to the B group are the same.

In this case, the feature information X _f is a value obtained by subtracting the sum of the projection data belonging to the group B from the sum of the projection data belonging to the group A as described in the equation (3).

The feature information generated for each frame is stored in the feature information storage unit 145, and a locus is generated using the feature information. And based on the generated locus, determines a relatively dominant axial direction through a mutual analysis (step 240) between the X and Y axes. The motion direction is determined by analyzing the feature information of the dominant axial direction (step 250). Here, if there is one reference axis, step 240 may be omitted. In step 250, the motion direction may be determined by analyzing the feature information in the direction of the reference axis.

The A group projection data sum 710a, the B group projection data sum 710b, and the feature information 720 in the first section, the second section and the third section are shown in FIG.

In the first section, the feature information on the X axis forms a sinusoidal waveform. On the other hand, the feature information on the Y-axis forms a waveform that can not be regarded as a sine form. That is, in the first section, the feature information on the X-axis constituting the sinusoidal waveform is judged to be relatively superior to the feature information on the Y-axis.

The characteristic information on the X axis has a positive value at the beginning and a negative value at the end, so that in the X axis, the first direction (for example, the direction in which x increases) (See D (1)).

In the second section, the characteristic information on the X axis forms a sinusoidal waveform. On the other hand, the feature information on the Y-axis forms a waveform that can not be regarded as a sign. That is, it is judged that the characteristic information on the X-axis forming the sinusoidal waveform in the second section is relatively superior to the characteristic information on the Y-axis.

The characteristic information on the X axis has a negative value at the beginning and a positive value having a positive value at a later time. In the second direction (for example, in the direction in which x decreases) in the X axis, (See D (2)).

In the third section, the motion of the subject can be determined through a process similar to that in the second section (see D (3)).

Each time the new feature information is sequentially stored for the feature information stored in the feature information storage unit 145, the following feature information waveform analysis method can be performed to determine whether the trajectory based on the current feature information is a sign feature . A method of analyzing the waveform of feature information according to an embodiment of the present invention is as follows.

It is assumed that the feature information has been extracted by performing steps 200 to 230 described above. A plurality of feature information is stored in the feature information storage unit 145 as a predetermined time elapses.

And characterized by the information based on the current frame for the feature information stored in the storage unit 145 using the characteristic information is located within a range of positive-characteristic value (X _pos), the negative characteristic value (X _neg), the direction characteristic value (X _dir ).

For example, when the nth frame image is input, it is assumed that a range from the frame before the ath frame to the frame before the bth frame is within a certain range based on the current frame. In this case, a value obtained by summing only the value of a positive characteristic value (X _pos) is the amount of the characteristic information is located within a range of (+) as described in equation (4), the negative characteristic value (X _neg) is located within a predetermined range (-) values of the feature information. The direction feature value (X _dir ) is a value obtained by summing differences between consecutive feature information within a certain range.

A frame when the positive feature value, the negative feature value, and the direction feature value respectively are equal to or greater than a preset threshold value is set as a start frame. When each feature value is equal to or greater than a preset threshold value, it can be determined that the feature information has a sign shape.

The positive feature value, the negative feature value, and the direction feature value according to Equation (4) are calculated from the start frame to a predetermined number of frames. For example, if the kth frame is determined as the start frame and the predetermined number is 5, the positive feature value, the negative feature value, and the direction feature value according to Equation (4) are calculated from the kth frame to the (k + 4) th frame.

The positive feature value and the negative feature value calculated from the start frame to a predetermined number of frames are respectively added. The direction characteristic value is divided into a positive direction characteristic value and a negative direction characteristic value by dividing a case having positive (+) value and a case having negative (negative) value.

After the above-described process is performed, feature values based on feature information on the X-axis and feature values based on feature information on the Y-axis are compared. The positive feature value is compared with the negative feature value, and the axis having the larger absolute value is determined as the main direction axis of the subject movement.

For the motion direction on the axis, the absolute value of the positive direction characteristic value is compared with the absolute value of the negative direction characteristic value. When the absolute value of the positive direction characteristic value is larger in the first direction (for example, in the direction of increasing x in the case of the X-axis), when the absolute value of the negative direction characteristic value is larger, For example, the direction in which x decreases in the case of the X axis).

On the other hand, the above-described motion detection method can be created by a computer program. The codes and code segments that make up the program can be easily deduced by a computer programmer in the field. The program is stored in a computer-readable medium and readable and executed by a computer to implement a motion detection method. The information storage medium includes a magnetic recording medium, an optical recording medium, and a carrier wave medium.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention as defined in the appended claims. It will be understood that the invention may be varied and varied without departing from the scope of the invention.

1 is a block diagram of a motion detection apparatus according to an embodiment of the present invention;

2 is a flowchart of a motion detection method according to an embodiment of the present invention.

3 is a diagram illustrating a method of generating a difference image;

Figure 4 is an illustration of a difference image.

Figure 5 shows an image projection in a difference image;

6 shows a method of extracting feature information.

7 shows a waveform of a feature information set;

Claims (20)

A motion detection method performed by a motion detection apparatus, Receiving a sequential frame image of a subject from an image sensor; Generating a difference image between a frame image of the input current frame and a frame image of a previous frame; Projecting the difference image onto a reference axis including a horizontal axis and a vertical axis of the difference image to obtain projection data; Extracting feature information from the projection data; Analyzing a trajectory of the feature information; And And acquiring motion information of the subject using the analyzed locus, In the step of extracting the feature information, the projection data is grouped based on a detection reference projection point, and the feature information is generated using the difference of the sum of projection data within each group divided on the basis of the detection reference projection point , In the step of analyzing the locus, it is determined whether or not the feature information accumulated and accumulated in order to determine the presence or absence of motion of the subject has a shape of a sine waveform set in advance over time, When the motion information of the subject is determined to be present, the trajectory of the feature information corresponding to the projection data projected on each of the horizontal axis and the vertical axis is relatively similar to the predetermined sinusoidal waveform Wherein one of the horizontal axis and the vertical axis is determined as the main direction of the motion of the subject. The method according to claim 1, Wherein the step of generating the difference image generates the difference image by using a difference between pixel values of pixels corresponding to each other between a frame image of the current frame and a frame image of the immediately preceding frame as pixel values. The method according to claim 1, Wherein the step of acquiring the projection data acquires the projection data by summing the pixel values of the pixels in the difference image with respect to the reference axis. The method according to claim 1, Wherein the step of acquiring the projection data acquires the number of pixels in the difference image with the pixel value being equal to or greater than a threshold value with respect to the reference axis as the projection data. delete delete delete delete delete A recording medium on which a program of instructions executable in a computer device for performing the motion detection method according to any one of claims 1 to 4 is recorded and a program that can be read by the computer device is recorded. An image input unit for receiving a sequential frame image of a subject from the image sensor; A difference image generation unit for generating a difference image between a frame image of a current frame and a frame image of a previous frame inputted by the image input unit; An image projection unit for projecting the difference image on a reference axis including a horizontal axis and a vertical axis of the difference image to obtain projection data; A feature information extracting unit that extracts feature information from the projection data; A trajectory analysis unit for analyzing a trajectory of the feature information; And And a motion analyzer for obtaining motion information of the subject using the analyzed locus, Wherein the feature information extracting unit groups the projection data around a detection reference projection point and generates the feature information by using a difference of a sum of projection data within each group divided based on the detection reference projection point, Wherein the trajectory analyzing unit determines whether or not the motion of the subject is determined by determining whether or not the accumulated feature information has a sine waveform shape set in advance over time, Wherein when the trajectory analyzing unit determines that the motion of the subject is present, the motion analyzing unit determines that the trajectory of the feature information corresponding to the projection data projected on each of the horizontal axis and the vertical axis is relatively similar to the predetermined sinusoidal waveform And the vertical axis is determined as the main direction of the motion of the subject. 12. The method of claim 11, Wherein the difference image generation unit generates the difference image in which a difference between pixel values of pixels corresponding to each other between a frame image of the current frame and a frame image of the immediately preceding frame is a pixel value. 12. The method of claim 11, Wherein the image projection unit obtains the projection data by summing pixel values of pixels in the difference image with respect to the reference axis. 12. The method of claim 11, Wherein the image projection unit obtains, as the projection data, the number of pixels in the difference image whose pixel value is greater than or equal to a threshold value with respect to the reference axis. delete delete delete delete delete 12. The method of claim 11, And an automatic exposure unit for adjusting an exposure value of the image sensor according to a surrounding environment so that the frame image is input at a constant frame rate.

Images