KR101536353B1 - A gesture recognition toy - Google Patents

Abstract

The present invention relates to a gesture recognition toy with an improved structure to enable the toy to recognize user′s commands at a distance. According to the present invention, the gesture recognition toy includes: a body part; a toy body part having a drive part to move the body part; an optical system which is shaped into a cone or a polygon to be installed on the body part and includes a reflective mirror to reflect an image in 360 degree direction around the body part into one direction; an image taking part to continuously take an image provided by the optical system; an image analysis part to analyze a continuous image taken by the image taking part; and a command transmission part to transmit a signal to drive the drive part in accordance with the analysis result from the image analysis part.

Description

A gesture recognition toy

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a motion recognition toy, and more particularly, to an operation recognition toy capable of performing an instruction by recognizing a user's specific motion.

With the continuous development of artificial intelligence technology, toys and robots equipped with voice recognition or motion recognition functions have been developed recently, and Japanese Laid-Open Patent Publication No. 10-2014-0054977 has been proposed. As such, artificial intelligence-equipped toys can give more intimacy to users, and it can be a useful system not only for children but also for adults and the elderly in modern households with one person.

On the other hand, in order for a toy to have the same level of intimacy as a peddler, it is important that the user accesses the user in front of the user. In order to implement such a function, the toy must be able to be called from a distance. However, speech recognition works relatively well when noise is small at a short distance, but is difficult to operate when it is at a long distance. Even if speech recognition is possible, it is difficult to track the user because it is difficult to grasp the position of the call with voice alone.

On the other hand, the motion recognition method can operate at a relatively long distance and can grasp the position where the motion occurred. However, in the case of motion recognition, since the image input to the image capturing unit installed in the toy is interpreted and operated, there is a problem in that the user can recognize the operation only when the user is in front of the image capturing unit.

Korean Patent Publication No. 10-2014-0054977

It is an object of the present invention to provide a motion recognition toy having a structure improved to allow a toy to recognize a command of a user even at a long distance.

A motion recognition toy according to the present invention includes a body having a body having a driving part for moving the body, a cone or a polygonal pyramid, and mounted on the body, An image capturing unit for continuously capturing an image provided through the optical system; an image analyzing unit for analyzing a continuous image captured by the image capturing unit; And a command transmitting unit for transmitting a signal to drive the driving unit according to the analyzed result.

According to the present invention, it is preferable that an angular velocity sensor for measuring the amount of rotation of the body part is installed on the body part.

According to the present invention, the image capturing unit, the image analyzing unit, and the command transmitting unit may be detachably mounted on the toy body, which is one independent device, It is preferable that a control unit is provided.

According to the present invention, it is preferable that the independent device is a smart phone.

According to the present invention, it is preferable that the toy main body is provided with fixing means for fixing the smartphone so that the image capturing unit of the smart phone is positioned at a position where the image provided through the optical system can be captured.

In addition, according to the present invention, the fixing means includes a first support member coupled to the body portion so as to be vertically movable with respect to the optical system, and a second support member coupled to the body portion to be movable in the left- Preferably, the smartphone is supported and fixed by the first support and the second support.

According to the present invention, the image analyzing unit obtains a difference image between the consecutive images, analyzes the obtained difference image to obtain a motion trajectory of an object in which the motion occurs, and calculates the obtained motion trajectory based on a predetermined standard trajectory . ≪ / RTI >

According to another aspect of the present invention, there is provided a path forming unit for confirming a position of a recognition mark having a predetermined shape in the photographed image and determining a relative position between the recognition mark and the body part to generate a movement path of the body part .

According to the invention, the toy can recognize the user's command even if the user is far from the toy and the user is in a different position than the toy.

1 is a schematic diagram of a motion recognition toy according to an embodiment of the present invention.
FIG. 2 is a flowchart for explaining image capturing and analysis and a process of controlling a driving unit accordingly.
3 is an image photograph taken at the image photographing unit.
4 is a diagram showing the movement of the user's hand.
5 is an image of an actual user.
FIG. 6 is a difference image between two images shown in FIG.
7 is a schematic block diagram of a motion recognition toy according to another embodiment of the present invention.
FIG. 8 is a flowchart for explaining image capturing and analysis, and a process of controlling a driving unit accordingly.
9 is a block diagram of a motion recognition toy according to another embodiment of the present invention.
10 is a view showing a recognition mark.

Hereinafter, a motion recognition toy according to a preferred embodiment of the present invention will be described with reference to the accompanying drawings.

FIG. 1 is a schematic diagram of a motion recognition toy according to an embodiment of the present invention. FIG. 2 is a flowchart for explaining image capturing and analysis and a process of controlling a driving unit according to an embodiment of the present invention. FIG. 4 is a view showing a hand movement of a user, FIG. 5 is an image of a real user, and FIG. 6 is a difference image between two images shown in FIG.

1 to 5, the motion recognition toy 100 according to the present embodiment includes a toy body 10, an optical system 20, an image pickup unit 30, an image analysis unit 40, And an instruction transfer unit 50. [

The toy body 10 is composed of a body part 11 and a driving part. 1, the body part 11 is a rectangle. However, the body part 11 may be a robot, a doll (a dog, a person, etc.) Shape. The driving part is for moving the body part 11, and may include a wheel 12 mounted on the body part and a motor (not shown) connected to the wheel. Further, the toy body 10 is provided with an angular velocity sensor (not shown) for measuring the angle at which the body is rotated (angle of rotation about the vertical axis).

The optical system 20 is configured to allow the image capturing unit 30 to capture an image in a 360 ° (360 ° around the vertical axis) centering on the body part 11, more precisely the optical system 20 provided in the body part . As described in the background art, in the case of directly capturing a user to the image capturing unit and recognizing the operation, the user must be positioned in front of the image capturing unit. To solve this problem, the optical system 20 is applied in the present invention. As shown in FIG. 1, the optical system 20 includes a reflection mirror 21 and an auxiliary mirror 22. The reflection mirror 21 is formed in a conical shape so that a 360 ° image (peripheral view) around the body is reflected by the reflection mirror 21 and then transmitted downward. The auxiliary mirror 22 is disposed at an angle of 45 degrees below the reflection mirror 21 so that the image reflected by the reflection mirror 21 is reflected by the auxiliary mirror 22 and then transmitted to the left. For reference, since a visible image is light reflected from a corresponding object, it is easy to understand if the arrow shown in Fig. 1 is regarded as light.

The image capturing unit 30 continuously captures an image provided by the optical system 20. [ More specifically, the image capturing unit 30 is disposed in front of the auxiliary mirror 22, and continuously captures an image formed on the auxiliary mirror 22. 3, it can be seen that a 360 ° (i.e., all directions) image is photographed with the optical system 20 as the image captured by the image capturing unit.

The image analyzing unit 40 analyzes a continuous image photographed by the image photographing unit 30 to analyze a motion locus of an object in which motion occurs, for example, for analyzing a user's motion. More specifically, the image analysis unit 40 obtains a difference image between consecutive images, and analyzes the obtained difference images to obtain a user's motion trajectory. For example, when the user shown in FIG. 4 shakes both hands, the position of the hand varies between successive captured images, and thus a difference image of the hand position can be obtained. 5 (A) and 5 (B), the positions of both hands of the user are different. Accordingly, a difference image between the images of FIGS. 5 (A) and 5 , A difference image as shown in Fig. 6 can be obtained. For reference, the portion indicated by white in FIG. 6 represents the difference image (i.e., the portion where the difference between the two images). And, by analyzing the car image in this way, it is possible to analyze the motion trajectory of the user, that is, the trajectory of the movement of the hand.

Then, in the image analysis unit 40, the motion locus obtained as described above is compared with a predetermined standard locus (that is, a predetermined hand motion of the user's hand) to determine whether the two loci coincide with each other.

The command transmission unit 50 transmits a signal to the driving unit so that the driving unit is driven according to the result analyzed by the image analysis unit 40. For example, if the image analysis results show that the user is shaking both hands, and the shaking motion of both hands is determined to move the toy body toward the user (i.e., a predetermined standard locus), the command transmission unit 50 Controls the driving unit so that the toy body 10 moves toward the user.

For reference, the position of the movement (i.e., the position of the user) in the difference image is checked, and when the toy body is rotated so as to face the position, the user can move toward the user by linearly moving the angular velocity sensor, An angle can be measured to determine the direction in which the toy body moves (i.e., whether it is in a direction of moving toward the user).

As described above, according to the present invention, since the optical system is provided, images of all directions can be taken on the basis of the toy body, and the user's commands can be confirmed by analyzing the images. Thus, the toy can be operated even if the user is located other than the front of the toy.

FIG. 7 is a schematic block diagram of a motion recognition toy according to another embodiment of the present invention, and FIG. 8 is a flowchart illustrating a process of imaging and analyzing and controlling a driving unit accordingly.

7 and 8, the motion recognition toy 200 according to the present embodiment includes a toy body 210, an optical system 220, an image capturing section 231, (232), and an instruction transfer unit (233). The structure or function of each structure is the same as the above-described embodiment, and the following description will focus on the difference between the present embodiment and the previous embodiment.

First, according to the present embodiment, the optical system 220 is coupled to the toy body 210 to form a single product. The image capturing unit 231, the image analyzing unit 232, and the command transmitting unit 233 are composed of one independent device, and are detachably mounted on the toy body. Preferably, the smartphone 230 may be applied to the independent device. That is, a camera of the smart phone 230 is used as the image capturing unit 231, and an AP (Application Processor) serving as a central processing unit of the smart phone 230 is connected to the image analyzing unit 232 and the command transmitting unit 233, . More specifically, if a smart phone is mounted on a toy body after installing an application capable of implementing functions of an image analysis unit and an instruction transfer unit on a smart phone, a camera provided in the smart phone captures an image, Analyzing it through an application, and outputting a signal according to the analysis result to the driving unit.

In order to receive a signal output from the smartphone 230, a controller having a communication function is preferably provided in the toy body. In this case, a wireless communication method such as Wi-Fi may be applied to the communication method, And can also be implemented in a wired connection to the phone.

According to the present invention, after an application is installed on a smart phone of a user (or a smart phone that is not used), the toy can be mounted on the toy body and used. In this case, since the camera and the electronic chip for image processing (i.e., functions for the image analysis unit and the command transfer unit) do not need to be embedded in the motion recognition toy, the manufacturing cost of the motion recognition toy can be reduced. Particularly, in order to analyze a user's motion, a high-performance camera is required. In order to analyze the image, a high-performance electronic chip is required. In this embodiment, Can be greatly reduced.

9 is a block diagram of a motion recognition toy according to another embodiment of the present invention.

Referring to FIG. 9, the motion recognition toy 300 according to the present embodiment further includes fixing means. The fixing means is for fixing the smartphone 330 to the toy body, in which the camera of the smartphone is fixed at a position where the image provided through the optical system 320 can be photographed. The fixing means according to the present embodiment includes a first support 312 and a second support 313. 9, the toy body 310 is provided with a fixing member 311 for fixing the optical system 320 to the toy body 310. As shown in FIG. The first support base 312 is vertically slidably coupled to the fixing member 311 so that the first support base 312 is moved up and down with respect to the optical system 320. A stopper member 3121 is provided at an end of the first support member. The stopper member 3121 is preferably made of a material having elasticity, for example, rubber. The second support base 313 is slidably coupled to the first support base 312 in a lateral direction so that the second support base 312 is moved in the left and right direction with respect to the optical system 320.

The first support base 312 is lowered downward as much as possible, and then the smartphone 330 is placed on the first support base 312 , The camera of the smart phone then raises the first support 312 to a position facing the optical system 320 and then fixes the first support 312. The lower end of the smartphone is brought into close contact with the stopper member 3121 until the camera is moved to the right until the camera is close to the optical system 320, And is fixed between the second support 3121 and the second support 313.

As described above, according to the present embodiment, smartphones of various sizes can be mounted on the toy body and used.

Meanwhile, the motion recognition toy may further include a path forming unit (not shown). The path forming part is for forming a movement path of the motion recognizing toy when the motion recognizing toy automatically moves to a specific position (i.e., the user does not operate the driving part but moves himself). To this end, the perimeter (wall or floor, etc.) of the motion recognition toy is firstly affixed with a recognition mark of a preset shape. At this time, it is preferable that the recognition mark is easily distinguishable from the surroundings. For example, as shown in Fig. 10, it is easily distinguishable from a plurality of different color combinations.

The path forming unit identifies the position of the recognition mark on the photographed image, and thereby grasps the relative position (distance and direction) between the motion recognition toy and the recognition mark. If the position (coordinate) of the recognition mark is stored in advance, it is possible to confirm the current position of the motion recognition toy. Then, in consideration of the current position and the position of the destination, a motion path (direction, distance, etc.) through which the motion recognition toy moves is formed.

For reference, the motion recognition toy is loaded with a battery for supplying power to the driving unit, and thus it is necessary to charge the battery regularly (or at the time of battery discharge). At this time, the path forming unit can confirm the position of the current motion recognition toy and form a movement path to the charger, so that the motion recognition toy can be automatically moved to the charger to perform charging.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is clearly understood that the same is by way of illustration and example only and is not to be taken by way of limitation in the embodiment in which said invention is directed. It will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the appended claims.

For example, the toy main body may be provided with an infrared ray output unit for outputting signals for controlling household appliances such as TVs, air conditioner lights, and the like. Then, the user can move the motion recognition toy to a nearby location of the user, and then operate the home appliance through an instruction (for example, a preset operation).

The toy body may further include a rotation driving unit for rotating the optical system and the smartphone. When the rotation driving unit is installed as described above, the user can rotate the smartphone so that the user can see the screen of the smartphone even when the smartphone is mounted on the toy body.

100 ... motion recognition toy 10 ... toy body
11 ... body part 12 ... wheels
20 ... optical system 21 ... reflection mirror
22 ... auxiliary mirror 30 ... image taking section
40 ... image analysis unit 50 ... command delivery unit

Claims (8)

A toy body having a body part and a driving part for moving the body part;
An optical system formed in a cone or polygonal pyramid shape and provided on the body part and including a reflection mirror for reflecting an image in a 360 ° direction in one direction about the body part;
Fixing means for fixing the smartphone to the toy body such that the image pickup unit of the smartphone is positioned at a position where the image provided through the optical system can be captured; And
And a controller for transmitting a signal to the smartphone,
Wherein the fixing means includes a first support which is coupled to the body so as to be movable up and down with respect to the optical system and a second support which is coupled to the body so as to be movable in the left and right direction with respect to the optical system, The smartphone is supported and fixed by the first support and the second support,
Wherein the smartphone continuously captures an image provided through the optical system, analyzes the captured continuous image, and transmits a signal to the controller to drive the driving unit according to the analyzed result. toy. The method according to claim 1,
Wherein the body part is provided with an angular velocity sensor for measuring the amount of rotation of the body part. delete delete delete delete The method according to claim 1,
Wherein the image analyzing unit obtains a difference image between the consecutive images,
The movement image of the object in which motion occurs is obtained by analyzing the obtained difference image,
And comparing the obtained motion locus with a predetermined standard locus. The method according to claim 1,
Further comprising a path forming unit for confirming a position of a recognition mark of a predetermined shape in the photographed image and for determining a relative position between the recognition mark and the body part to generate a movement path of the body part, Motion recognition toys.

Images