1. A computer vision recognition device comprising a palm member (201) and a rubber finger tie (202), characterized in that: five holes corresponding to the rubber finger knots (202) are formed in the palm piece (201), and the five rubber finger knots (202) are all inserted into the palm piece (201).

2. A computer vision recognition device according to claim 1, wherein: the rubber finger joint is characterized by further comprising a rotating finger joint (203), a rotating shaft (204) and a sleeve (205), the sleeve (205) is inserted into the upper portion of each rubber finger joint (202), the rotating finger joint (203) is hinged to each sleeve (205) through the rotating shaft (204), and a rubber sleeve is sleeved on the rotating shaft (204).

3. A computer vision recognition device according to claim 2, wherein: the palm piece is characterized by further comprising a socket (106), and the lower portion of the palm piece (201) is plugged into the socket (106).

4. A computer vision recognition device according to claim 3, wherein: the socket is characterized by further comprising a round seat (101), extending pieces (104) and extension springs (105), wherein the four extending pieces (104) are annularly arranged on the periphery of the round seat (101), the extension springs (105) are connected to the four extending pieces (104), and the other ends of the four extension springs (105) are respectively connected to the periphery of the socket (106).

5. A computer vision recognition device according to claim 4, wherein: the circular seat is characterized by further comprising a motor I (102) and a sliding piece (103), wherein the sliding piece (103) is fixedly connected with the motor I (102), and the upper portion of an output shaft of the motor I (102) is fixedly connected to the center of the lower side of the circular seat (101).

6. A computer vision recognition device according to claim 5, wherein: the sliding piece (103) and the parts on the upper side of the sliding piece (103) are respectively arranged on the left and the right.

7. A computer vision recognition device as defined in claim 6, wherein: still include projection (107), T shape frame (301), reach pole (304), slot hole (305), lever (306) and motor II (307), respectively be provided with one and reach pole (304) before about on T shape frame (301), two slider (103) sliding connection respectively are on two reach pole (304), the middle part fixedly connected with motor II (307) of T shape frame (301), the output shaft fixed connection of motor II (307) is in the middle part of lever (306), two portions all are provided with slot hole (305) about lever (306), the equal fixedly connected with projection (107) of downside of two slider (103), two projection (107) sliding connection respectively are on two slot holes (305).

8. A computer vision recognition device according to claim 7, wherein: still include telescopic link I (302) and sliding member (303), the equal sliding connection in both ends has sliding member (303) about T shape frame (301), the rear portion of two preceding poles (304) fixed connection respectively is in the front side of two sliding member (303), the middle part fixedly connected with two telescopic link I (302) of T shape frame (301), the outer end of two telescopic link I (302) is fixed connection respectively on two sliding member (303).

9. A computer vision recognition device according to claim 8, wherein: the T-shaped frame is characterized by further comprising a cross bottom frame (601) and a supporting plate (604), the supporting plate (604) is fixedly connected to the lower side of the T-shaped frame (301), and the cross bottom frame (601) is fixedly connected to the lower portion of the supporting plate (604).

10. A computer vision recognition device according to claim 9, wherein: the novel cross chassis comprises a cross chassis (601), and is characterized by further comprising a vertical rod (602) and a camera arranging plate (603), wherein the vertical rod (602) is fixedly connected to the front part of the cross chassis (601), and the camera arranging plate (603) is fixedly connected to the upper part of the vertical rod (602).

Background

Gesture recognition in computer vision is a technology for recognizing human body gestures through a mathematical algorithm, and the human body gestures can be recognized by three-dimensionally imaging a sensor of a camera. 202021916493.0, which comprises a box body, a fixed platform fixed on one side of the box body, and a recognition camera rotatably connected to the middle part of the fixed platform, wherein the utility model comprises a fixed column, a joint column and a fixed box, an operator pulls the fixed column to drive the joint column to slide in the inner cavity of the fixed box, the bottom end of the joint column is separated from the joint groove, a thrust spring pushes the fixed plate to move, the fixed plate moves to a position suitable for maintenance of the operator, the operator releases a handle, the fixed spring pushes the joint column to slide downwards along the inner cavity of the fixed box, the bottom end of the joint column extends into the joint groove to fix the fixed plate, and the operator can conveniently maintain a receiving module, a processing module and a display output module, the use experience of the operators is improved. However, the patent can not form a plurality of gestures for computer vision recognition through the palm piece and the rubber finger knots.

Disclosure of Invention

In order to overcome the defects of the prior art, the invention provides the computer vision recognition device which has the beneficial effect that the invention can form various gestures for computer vision recognition through the palm component and the rubber finger joints.

A computer vision recognition device comprises a palm piece and rubber finger knots, wherein five holes corresponding to the rubber finger knots are formed in the palm piece, and the five rubber finger knots are all inserted in the palm piece.

The computer vision recognition device further comprises a rotating finger knot, a rotating shaft and a suite, wherein the suite is inserted into the upper part of each rubber finger knot, the rotating finger knot is hinged to each suite through the rotating shaft, and the rotating shaft is sleeved with a rubber sleeve.

The computer vision recognition device also comprises a socket, and the lower part of the palm piece is plugged in the socket.

The utility model provides a computer vision recognition device still includes circle seat, overhanging piece and extension spring, and the periphery of circle seat is the annular and is equipped with four overhanging pieces, all is connected with extension spring on four overhanging pieces, and four extension spring's the other end is connected respectively around the socket.

The computer vision recognition device further comprises a motor I and a sliding piece, wherein the sliding piece is fixedly connected with the motor I, and the upper part of an output shaft of the motor I is fixedly connected to the center of the lower side of the round seat.

The sliding part and the part on the upper side of the sliding part are respectively provided with one part at the left and the right.

The utility model provides a computer vision recognition device still includes the projection, T shape frame, the pole that stretches out before, the slot hole, lever and motor II, respectively be equipped with the pole that stretches out before about on the T shape frame, two sliders are sliding connection respectively on two the pole that stretches out before, the middle part of T shape frame is equipped with motor II, motor II's output shaft is established at the middle part of lever, two portions all are equipped with the slot hole about the lever, the downside of two sliders all is equipped with the projection, two projections sliding connection respectively are on two slot holes.

The utility model provides a computer vision recognition device still includes telescopic link I and slides, and the equal sliding connection in both ends has the piece of sliding about the T shape frame, and the rear portion of stretching the pole before two establishes respectively in the front side of two pieces of sliding, and the middle part of T shape frame is equipped with two telescopic link I, and two telescopic link I's outer end is established respectively on two pieces of sliding.

The computer vision recognition device further comprises a cross underframe and a supporting plate, the supporting plate is arranged on the lower side of the T-shaped frame, and the cross underframe is arranged on the lower portion of the supporting plate.

The computer vision recognition device further comprises an upright rod and a camera mounting plate, the upright rod is arranged at the front part of the cross-shaped underframe, and the camera mounting plate is arranged at the upper part of the upright rod.

The computer vision recognition device has the advantages that:

the invention can form a plurality of gestures for computer visual recognition through the palm component and the rubber finger knots.

Drawings

The invention is described in further detail below with reference to the accompanying drawings and specific embodiments.

FIG. 1 is a first schematic structural diagram of a computer vision recognition device;

FIG. 2 is a schematic structural diagram II of the computer vision recognition device;

FIG. 3 is a third schematic structural diagram of a computer vision recognition device;

FIG. 4 is a fourth schematic structural diagram of a computer vision recognition device;

FIG. 5 is a schematic structural diagram of a computer vision recognition device;

FIG. 6 is a first schematic structural view of the round seat and the palm member;

FIG. 7 is a second schematic structural view of the round seat and the palm member;

FIG. 8 is a schematic structural view of a T-shaped frame;

FIG. 9 is a first schematic view of a ring;

FIG. 10 is a second schematic view of the ring structure;

FIG. 11 is a schematic view of the construction of the cowl;

fig. 12 is a schematic structural view of a cross chassis.

In the figure: a circular base 101; a motor I102; a slide member 103; an overhang 104; an extension spring 105; a socket 106; a convex column 107;

a palm member 201; a rubber finger tie 202; a rotating finger knot 203; a rotating shaft 204; a kit 205;

a T-shaped frame 301; a telescopic rod I302; a glide 303; a forward rod 304; a long hole 305; a lever 306; motor II 307;

a circular ring 401; an elastic plate 402; a vertical slide bar 403; a boss 404; a telescopic rod II 405; motor III 406; an L-shaped frame 407; a beam rail 408;

an arc-shaped baffle 501; an empty slot 502; a stopper 503; a stopper 504; a compression spring 505;

a cross chassis 601; a vertical rod 602; a camera mounting plate 603; a support plate 604.

Detailed Description

Referring to fig. 1-7, an embodiment of the present invention is provided to solve the main technical problem of inserting different numbers of rubber finger joints 202 to form gestures for computer recognition, testing the gesture recognition capability of the computer, and further,

because the computer vision recognition device comprises the palm piece 201 and the rubber finger knots 202, five holes corresponding to the rubber finger knots 202 are formed in the palm piece 201 and are respectively used for inserting the five rubber finger knots 202, the five rubber finger knots 202 simulate human fingers, and then different numbers of rubber finger knots 202 are inserted according to requirements to form gestures for recognition of a computer, so that the gesture recognition capability of the computer is tested.

Referring to fig. 1-7, an embodiment of the present invention is provided, which solves the main technical problem of providing more gestures for a computer to recognize, testing the recognition capability of the computer for various gestures, and further,

because the computer vision recognition device further comprises the rotating finger knots 203, the rotating shaft 204 and the suite 205, the suite 205 is sleeved on the upper portion of each rubber finger knot 202, the suite 205 plays a role in connection between the rotating finger knots 203 and the rubber finger knots 202, the rotating finger knots 203 can rotate on the rubber finger knots 202 through the rotating shaft 204, the joint positions of human fingers are further simulated, manual setting of more gestures is further completed, more gestures are provided for recognition of a computer, and the recognition capability of the computer on various gestures is tested. Because the rotating shaft 204 is sleeved with the rubber sleeve, the friction force between the rotating finger knot 203 and the rubber finger knot 202 is further increased, so that the relative positions of the rotating finger knot 203 and the rubber finger knot 202 can be fixed after the angles of the two are adjusted.

Referring to fig. 6, an embodiment of the present invention is provided, which solves the main technical problem of enabling the palm component 201 to be erected for computer recognition, and further,

since the computer vision recognition device further comprises the socket 106, the lower part of the palm member 201 is inserted into the socket 106, and the socket 106 supports the palm member 201, so that the palm member 201 can be erected for computer recognition.

Referring to fig. 6-7, an embodiment of the present invention is provided to test the gesture recognition effect of a computer under the condition that the gesture is continuously jittered, and further,

because computer vision recognition device still includes circle seat 101, overhanging piece 104 and extension spring 105, the periphery of circle seat 101 is provided with four overhanging pieces 104, all be provided with extension spring 105 on four overhanging pieces 104, four extension spring 105's the other end is connected respectively around socket 106, and then socket 106 is supported through the pulling of four extension spring 105 at this moment, four extension spring 105 are not stable enough to socket 106's support, and then make socket 106 rock easily, and then make palm piece 201 and rubber indicate knot 202 also rock easily, the position of people's hand is not fixed in reality, and then simulate dynamic staff, test the gesture recognition effect of computer under the condition that the gesture constantly trembles.

Referring to fig. 1-7, an embodiment of the present invention is provided to solve the main technical problem of recognizing a gesture under a gesture rotation condition, testing gesture recognition effect under more conditions, and further,

because the computer vision recognition device further comprises a motor I102 and a sliding part 103, the motor I102 is arranged on the sliding part 103, an output shaft of the motor I102 is fixedly connected to the round base 101, the sliding part 103 can be driven to rotate by taking the axis of the output shaft of the motor I102 as a shaft through the motor I102, the palm part 201 and the rubber finger joint 202 are driven to rotate by taking the axis of the output shaft of the motor I102 as a shaft, the rotating position of the gesture is changed, the gesture is recognized under the gesture rotating condition, and the gesture recognition effect under more conditions is tested.

Referring to fig. 1-8, an embodiment of the present invention is provided to solve the main technical problem of enabling a computer to recognize gestures of both hands of a person, and further,

the sliding part 103 and the part on the upper side of the sliding part 103 are respectively provided with one part on the left and the right, so that the gestures of both hands of a person are simulated, and the computer can recognize the gestures of both hands of the person.

Referring to fig. 1-12, an embodiment of the present invention is provided, which solves the main technical problem of making two gestures move in opposite directions with equal distance, in this case, testing the effect of computer recognition gesture, and further,

because the computer visual identification device also comprises convex columns 107, T-shaped frames 301, forward extending rods 304, long holes 305, levers 306 and motors II307, the left and the right of the T-shaped frames 301 are respectively provided with one forward extending rod 304, the two forward extending rods 304 are both connected with the sliding parts 103 in a sliding way, the middle part of the T-shaped frames 301 is provided with the motors II307, the output shaft of the motors II307 is fixedly connected with the middle part of the levers 306, the levers 306 can be driven to rotate when the motors II307 rotate, the left and the right ends of the levers 306 are both provided with the long holes 305, the convex columns 107 at the lower sides of the two sliding parts 103 are respectively connected with the two long holes 305 in a sliding way, further, the motor II307 drives the lever 306 to rotate and drives the two convex columns 107 to move in opposite directions with equal distance, thereby driving the two sliding members 103 to move in opposite directions at equal distances on the two front rods 304, and then the two gestures are moved in opposite directions with equal distance, and in this case, the effect of the computer on recognizing the gestures is tested.

Referring to fig. 1-12, an embodiment of the present invention is provided, which solves the main technical problem of testing the effect of computer gesture recognition under the condition that two hands are at different distances, and further,

because computer visual identification device still includes telescopic link I302 and sliding member 303, the equal sliding connection in both ends of T shape frame 301 has sliding member 303, and the front side at two sliding member 303 is connected respectively to the rear portion of two preceding poles 304, and the middle part of T shape frame 301 is provided with two telescopic link I302, and the flexible end of two telescopic link I302 is fixed connection respectively on two sliding member 303. And then can drive two sliding pieces 303 respectively through the flexible of two telescopic links I302 and move about on T-shaped frame 301, and then drive two preceding poles 304 and move about, can drive the left and right positions of two sliders 103 when two preceding poles 304 move about, and then change the interval between two palm pieces 201, test the effect that the computer discerned the gesture under the condition that two hands are in different intervals.

Referring to fig. 1-12, an embodiment of the present invention is provided, which solves the main technical problem that a cross bottom frame 601 and a supporting plate 604 are used to support a T-shaped frame 301, and further,

since the computer vision recognition apparatus further includes a cross chassis 601 and a support plate 604, the support plate 604 is provided at the lower side of the T-shaped frame 301, and the cross chassis 601 is provided at the lower portion of the support plate 604. Both the cross chassis 601 and the support plate 604 are used to support the T-shaped frame 301.

Referring to fig. 1-12, an embodiment of the present invention is provided, which solves the main technical problem of arranging a camera for computer vision recognition on a camera placing board 603, and further,

because the computer vision recognition device further comprises an upright 602 and a camera placing plate 603, the upright 602 is arranged at the front part of the cross underframe 601, the camera placing plate 603 is arranged at the upper part of the upright 602, and the camera used for computer vision recognition is placed on the camera placing plate 603.

The computer vision recognition device further comprises a circular ring 401, a motor III406, an L-shaped frame 407, a beam rod 408 and an arc-shaped baffle 501, the lower portion of the L-shaped frame 407 is inserted into the sliding piece 303, the upper portion of the L-shaped frame 407 is fixedly connected with the motor III406, the upper side of the circular ring 401 is fixedly connected with the beam rod 408, an output shaft of the motor III406 is fixedly connected to the middle portion of the beam rod 408, and the edge of the lower side of the circular ring 401 is provided with the arc-shaped baffle 501.

Referring to fig. 1-12, an embodiment of the present invention is provided to solve the main technical problem of testing the effect of computer gesture visual recognition under the condition of randomly blocking the gesture, and further,

when the motor III406 rotates, the beam rod 408 and the ring 401 are driven to rotate by taking the axis of the output shaft of the motor III406 as a shaft, and then the arc-shaped baffle 501 is driven to rotate by taking the axis of the output shaft of the motor III406 as a shaft, because the arc-shaped baffle 501 is always positioned at a certain position around the palm piece 201, the gesture can be blocked when the arc-shaped baffle 501 rotates by taking the axis of the output shaft of the motor III406 as a shaft, and the effect of computer gesture visual identification is tested under the condition that the gesture is blocked randomly.

The computer vision recognition device further comprises an empty groove 502, a stop head 503, a stop block 504 and a compression spring 505, the empty groove 502 is formed in the arc-shaped baffle plate 501, the stop block 504 is vertically and slidably connected to the empty groove 502, the compression spring 505 is arranged between the upper side of the stop block 504 and the upper portion of the arc-shaped baffle plate 501, and the two stop heads 503 are fixedly connected to the lower side of the arc-shaped baffle plate 501.

Referring to fig. 1-12, an embodiment of the present invention is provided to test the effect of computer gesture visual recognition under the condition of randomly blocking the upper and lower positions of the gesture, and further,

the compression spring 505 always gives the stopper 504 a downward force, so that the stopper 504 always has a downward movement trend, the stopper 503 prevents the stopper 504 from being separated from the arc-shaped baffle 501, the stopper 504 can randomly block the upper position and the lower position of a gesture when vertically sliding on the empty groove 502, and then the effect of computer gesture visual identification is tested under the condition that the upper position and the lower position of the gesture are randomly blocked.

The computer vision recognition device further comprises an elastic plate 402, a vertical sliding rod 403, a convex seat 404 and an expansion rod II405, the convex seat 404 is fixedly connected to the circular ring 401, the convex seat 404 is located at the arc-shaped baffle 501, the vertical sliding rod 403 is connected to the convex seat 404 in a vertical sliding mode, the elastic plate 402 is fixedly connected to the lower side of the vertical sliding rod 403, the expansion rod II405 is fixedly connected to the convex seat 404, the upper portion of the expansion rod II405 is fixedly connected to the upper portion of the vertical sliding rod 403, and the elastic plate 402 can move to the lower side of the stop block 504.

Referring to fig. 1-12, an embodiment of the present invention is provided to test the effect of computer gesture visual recognition under the condition of randomly blocking the upper and lower positions of the gesture, and further,

the telescopic link II405 drives the vertical sliding rod 403 and the elastic plate 402 to vertically slide when extending and retracting, when the elastic plate 402 moves to the lower side of the stop block 504, the elastic plate 402 rises to lift the stop block 504, so that the stop block 504 moves upwards relative to the arc-shaped baffle 501, at the moment, the compression spring 505 is compressed, when the elastic plate 402 cannot bear the elastic force of the compression spring 505, the stop block 504 rapidly launches downwards on the empty groove 502, further, the stop block 504 randomly slides vertically in the empty groove 502, and the effect of testing the gesture visual identification of the computer under the condition of randomly stopping the upper position and the lower position of the gesture is realized.