CN101667059A - Gesture recognition-based wireless intelligent judgment system - Google Patents

Abstract

The wireless intelligent referee system based on gesture recognition, in the current arena, there is a non-negligible time interval between observing the referee's gesture with the naked eye and performing the corresponding operation, and for the recorder, the workload is heavy and it is easy to misuse , to some extent affected by subjective factors. The present invention comprises: a glove (1), a conductive rubber sheet (2) on the palm part of the front of the glove, grounded or zero potential after the conductive rubber sheet is connected in series with a resistor, conductive rubber (3) is provided on the fingertips of the five fingers of the glove, and Connected to the power supply on the back of the glove, the ring finger and the second joint of the middle finger each have a conductive rubber, which is grounded or zero potential after being connected in series with a resistor. A group of acceleration sensors (4) and a group of geomagnetic sensors (5) are installed on the back of the glove. The present invention is applied to basketball, volleyball, baseball, rugby, hockey, etc., as well as on the field similar to the aforementioned sports.

Description

Wireless Intelligent Referee System Based on Gesture Recognition

Technical field:

The invention relates to a wireless intelligent referee system based on gesture recognition.

Background technique:

The referee's ruling on the field determines the progress of a game, and the real-time and accurate communication and understanding of the referee's ruling is of great significance in a game. In the current arena, after the referee makes a referee gesture, the sideline recorder observes the referee's gesture, then records it and performs other corresponding operations. There is a non-negligible time interval between observing the referee's gesture with the naked eye and performing the corresponding operation. Moreover, for the recorder, the workload is heavy, and it is easy to misuse, which is affected by subjective factors to a certain extent. In addition, since the referee's actions are completed in an instant, this also greatly affects the scorer's recognition of the action.

Invention content:

The purpose of the present invention is to provide an intelligent gesture recognition system, which collects the gestures of referees through acceleration sensors and geomagnetic sensors, and can transmit the gestures of referees on the playing field back to the computer system in real time and accurately. After analysis and calculation, intelligent recognition is finally realized.

Above-mentioned purpose realizes by following technical scheme:

A wireless intelligent referee system based on gesture recognition, its composition includes: gloves, the palm part of the front of the glove has a conductive rubber sheet, the conductive rubber sheet is connected to a resistor in series and then grounded or zero potential, the fingers of the five fingers of the glove The tip parts each have conductive rubber, and are connected to the power supply on the back of the glove, and the second joints of the ring finger and middle finger have conductive rubber, which are connected to the resistor in series and then grounded or zero potential. The back of the glove is equipped with a set of acceleration sensors and a set of geomagnetic sensor.

In the wireless intelligent referee system based on gesture recognition, the acceleration sensor is a three-axis acceleration sensor, and the geomagnetic sensor is a two-axis geomagnetic sensor, wherein one geomagnetic sensor and the acceleration sensor are placed on the same plane, and the other geomagnetic sensor is placed on the same plane as the acceleration sensor. The sensor is placed on the surface perpendicular to the acceleration sensor, and the signal collected by the arm movement and the finger movement is sent to the receiving chip, and the receiving chip transmits it to the remote radio frequency receiving end, and the radio frequency receiving end is controlled by the receiving chip nRF24E1 and USB The interface is formed, and the received data is demodulated and decoded at the receiving end, and then transmitted to the computer through the USB port.

The wireless intelligent referee system based on gesture recognition uses geomagnetic sensors and acceleration sensors to recognize the movements of the fingers and arms of the carrier, and uses conductive rubber to recognize the flexion and extension states of the fingers of the carrier. Process and restore the motion trajectory of the carrier's arms and fingers for gesture recognition.

The wireless intelligent referee system based on gesture recognition, using Gesture Recognition, a Windows desktop program developed by VC++, can analyze and calculate the information imported through the USB port, and redraw the gesture trajectory of the referee, and finally achieve gesture recognition the goal of.

This technical solution has the following beneficial effects:

1. Intelligent identification and recording saves a lot of work of off-site recorders

In order to be able to record the referee's referee on the field accurately and in real time, the off-field recorder must always pay attention to the referee's actions, which is a very heavy task for the off-field recorder. In fact, in order to ensure that the recorder can record every referee action of the referee, at least one recorder is arranged in each arena. However, if this system is adopted, it will save part of the cost of hiring a scorer for the club. More importantly, the intelligent action recognition and recording of this system saves a lot of work for the scorer off the field, making the scorer only You need to turn on the computer, move the mouse, and press the button to complete the recording work, even if no one is recording.

2. Strong real-time performance, so that every fan can keep abreast of the progress of the game on the field

In some large-scale competitions, narrators will be arranged, and there is a long time delay between the recorder seeing the referee's action, making the record and then passing it to the narrator. Through the LED display attached to this system, the referee's ruling can be provided to the fans in text form, and then cooperate with the commentary of the off-court commentator, the effect will be better.

The chance of misidentification is small.

3. The present invention adopts the integral path and compares it with the posture of the standard database to recognize gestures. Compared with the naked eye observation of the recorder, the probability of misjudgment is smaller. It also avoids the impact of the recorder's subjective factors on the recognition results.

Rich interfaces for further data processing

4. data processing of the present invention is mainly carried out in computer, and computer has USB, serial port, parallel port, interfaces such as RJ45, therefore can send out data according to different needs, do further processing.

5. Good scalability

Because the interface of the present invention is rich, it can be connected to different peripherals or hosts, and has relatively good scalability. In addition, the present invention is also fully applicable to other situations involving gesture recognition, such as dumb language recognition, traffic police gesture recognition, and mobile phones. In the case of emergency dialing, etc., you only need to change the database of the recognition software.

Description of drawings:

Accompanying drawing 1 is a structural block diagram of a wireless intelligent referee system based on gesture recognition;

Accompanying drawing 2 is a schematic diagram of the front of the glove;

Accompanying drawing 3 is the schematic diagram of glove back side;

Accompanying drawing 4 is a structural diagram of conductive rubber;

Accompanying drawing 5 is to get 2 points gesture figure;

Accompanying drawing 6 is the placement diagram of the sensor on the back of the glove;

Accompanying drawing 7 is the relationship diagram of accelerometer placement surface and horizontal plane.

The specific embodiment of the present invention:

Example 1:

A wireless intelligent referee system based on gesture recognition, its composition includes: a glove 1, the palm part of the front of the glove has a conductive rubber sheet 2, the conductive rubber sheet is grounded or zero potential after being connected in series with a resistor, and the five fingers of the glove Each of the fingertips has a conductive rubber 3, and is connected to the power supply on the back of the glove. The second joint of the ring finger and the middle finger has a conductive rubber, which is grounded or zero potential after being connected in series with the resistor. A group of acceleration sensors are installed on the back of the glove. 4 and a group of geomagnetic sensors 5.

In the wireless intelligent referee system based on gesture recognition, the acceleration sensor is a three-axis acceleration sensor, and the geomagnetic sensor is a two-axis geomagnetic sensor, wherein one geomagnetic sensor and the acceleration sensor are placed on the same plane, and the other geomagnetic sensor is placed on the same plane as the acceleration sensor. The sensor is placed on the surface perpendicular to the acceleration sensor, and the signal collected by the arm movement and the finger movement is sent to the receiving chip, and the receiving chip transmits it to the remote radio frequency receiving end, and the radio frequency receiving end is controlled by the receiving chip nRF24E1 and USB The interface is formed, and the received data is demodulated and decoded at the receiving end, and then transmitted to the computer through the USB port.

The wireless intelligent referee system based on gesture recognition uses geomagnetic sensors and acceleration sensors to recognize the movements of the fingers and arms of the carrier, and uses conductive rubber to recognize the flexion and extension states of the fingers of the carrier. Process and restore the motion trajectory of the carrier's arms and fingers for gesture recognition.

The wireless intelligent referee system based on gesture recognition, using Gesture Recognition, a Windows desktop program developed by VC++, can analyze and calculate the information imported through the USB port, and redraw the gesture trajectory of the referee, and finally achieve gesture recognition the goal of.

Example 2:

According to the realized function, the glove is divided into two parts: finger motion acquisition and arm motion acquisition. The overall structure diagram is shown in Figure 1.

Finger motion acquisition part (as shown in Figure 2 and Figure 3):

The palm of the front of the glove has a large piece of conductive rubber, which is grounded or zero potential after being connected in series with a 5K resistor. There is a small piece of conductive rubber on the fingertips of the five fingers, which are connected to the +5V power supply on the back of the glove, and the second finger of the ring finger and middle finger There is a small piece of conductive rubber at the joint, as shown in Figure 2, and the ground or zero potential after the 5K resistor is connected in series, as shown in Figure 4.

Take the score of 2 points for the referee's gesture as an example. When the referee makes this gesture, the ring finger and little finger bend to touch the conductive rubber on the palm of the hand, and at the same time the thumb bends to touch the conductive rubber at the second joint of the ring finger, as shown in Figure 5, and the circuit is turned on , then the pin of the receiving chip receives a high level, so that the movement of the finger can be collected. The collection of other action states is similar to the above.

Arm motion collection part:

The collection of arm movements mainly relies on the acceleration sensor and geomagnetic sensor on the back of the glove. The structure is shown in Figure 6. The acceleration sensor is a three-axis acceleration sensor MXR9500G produced by MEMSIC, which can measure the three-dimensional acceleration of the bottom surface of the plane on which it is placed.

The geomagnetic sensor is a two-axis geomagnetic sensor MMC2021M produced by MEMSIC, and the positions of the two geomagnetic sensors are shown in the structure diagram. Send the arm movement and finger movement to the receiving chip nRF24E1 through the circuit shown in the figure, and transmit it to the remote RF receiving end.

RF receiver:

The RF receiving end is composed of receiving chip nRF24E1 and USB interface. At the receiving end nRF24E1 demodulates and decodes the received data, and then sends the signal to the USB interface, which is then transmitted to the computer.

computer:

In the computer, the present invention utilizes VC++ to send out a visualization program to process the data transmitted from the USB port. The processing process is as follows: the data of the acceleration sensor is converted first, and then the obtained acceleration is integrated twice to the time. The movement trajectory of the arm can be obtained, and then the obtained movement trajectory is compared with the standard movement trajectory for recognition.

The placement of the sensor on the back of the hand is shown in Figure 6. Since the acceleration sensor can only measure the acceleration of its placement plane and its vertical plane, that is, the acceleration along x', y', z' in Figure 7, in order to facilitate the data processing of the computer, the collected data must be transformed once.

The acceleration sensor of the referee's action is located on the back of the hand of the glove, and the acceleration data collected by it satisfies the function, so the function must be about four variables, namely

$\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; a</span>}<span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; ,</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}<span\; class="notranslate">\; ,</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; z</span>}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>$

Wherein refers to the three vectors that determine the three-dimensional space, the plane is the horizontal plane, is the vector perpendicular to the horizontal plane, and is the time variable.

的，而与传感器在该时刻所处的位置有关，即与

直接相关的，如何把

变成

由附图6可知，只要知道了α，β两角的值，便可以很容易得到：The plane where the acceleration sensor is located at a certain moment cannot always be horizontal, as shown in Figure 7, which means that the raw data output by the three-axis acceleration sensor is not about

, and it is related to the position of the sensor at that moment, that is, it is related to

directly related, how to put

become

It can be seen from Figure 6 that as long as the values of the two corners of α and β are known, it can be easily obtained:

$\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{{\mathrm{<span\; class="notranslate">\; x</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}$

$\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}<span\; class="notranslate">\; =</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; +</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{{\mathrm{<span\; class="notranslate">\; x</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}$

$\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{\mathrm{<span\; class="notranslate">\; z</span>}}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; (</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{{\mathrm{<span\; class="notranslate">\; x</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; -</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&alpha;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}<span\; class="notranslate">\; +</span>\stackrel{<span\; class="notranslate">\; \&Right\; Arrow;</span>}{{\mathrm{<span\; class="notranslate">\; z</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&beta;</span>}$

So how to find the value of α, β? This is what geomagnetic sensors do. In the present invention, we use two two-axis MMC202XMs to realize the positioning of the plane where the sensor is located, that is, to obtain the values of the two angles of α and β.

Place an MMC202XM on the x'oy' plane to measure the value of α, and place an MMC202XM on the x'oz' plane to measure the value of β.

We can get from the acceleration sensor The value of the instantaneous acceleration in the direction, the value of the angle α, β between the acceleration sensor and the horizontal plane and the plane where the acceleration sensor is obtained can be obtained through two geomagnetic sensors. We send these data back to the computer with radio frequency technology, then the computer can use The received data, combined with the formula derived above, can determine the acceleration of the arm at any time.

Assuming that the initial velocity of the arm is 0, and the acceleration of the arm at any time is known, the velocity function can be obtained by integrating the obtained acceleration function with respect to time, and the path function of the arm can be obtained by integrating the velocity function again. In this way, it can be compared and identified with the standard function input in the computer in advance, and the obtained path function can be discriminated by mathematical methods such as the least square method.

Claims (5)

1. wireless intelligent judgment system based on gesture identification, its composition comprises: gloves, it is characterized in that: the palm of the hand in described gloves front partly has conductive rubber slice, ground connection or zero potential after described conductive rubber slice is connected with resistance, the finger tip part of the described gloves the five fingers respectively has conductive rubber, and receive the power supply at the gloves back side, respectively there is conductive rubber at second joint place nameless and middle finger, ground connection or zero potential after connecting with resistance, one group of acceleration transducer and one group of geomagnetic sensor are equipped with in the described gloves back side.

2. the wireless intelligent judgment system based on gesture identification according to claim 1, it is characterized in that: described acceleration transducer is a 3-axis acceleration sensor, described geomagnetic sensor is the diaxon geomagnetic sensor, wherein plot Magnetic Sensor and acceleration transducer are placed at grade, the plot Magnetic Sensor is placed on the face vertical with acceleration transducer in addition, give receiving chip the action acquired signal of the action of arm and finger, be transmitted to the radio frequency receiving end of far-end by receiving chip, described radio frequency receiving end is made of receiving chip nRF24E1 and USB interface, carry out demodulation at receiving end receiving data, decoding is delivered in the computing machine by the USB oral instructions then.

3. the wireless intelligent judgment system based on gesture identification according to claim 1 and 2, it is characterized in that: the finger and the arm action that use geomagnetic sensor and acceleration transducer identification carrier, use the state that bends and stretches of conductive rubber identification carrier finger, carrier's the arm and the movement locus of finger are handled and reduced to the computing machine that information is reached far-end by transmitting chip, thereby carry out gesture identification.

4. the wireless intelligent judgment system based on gesture identification according to claim 1 and 2, it is characterized in that: utilize Gesture Recognition, the windows desktop program of VC++ exploitation, can carry out analytical calculation to the information of going into through the USB oral instructions, and the referee's of refigure place gesture track, finally reach the purpose of gesture identification.

5. the wireless intelligent judgment system based on gesture identification according to claim 3, it is characterized in that: utilize Gesture Recognition, the windows desktop program of VC++ exploitation, can carry out analytical calculation to the information of going into through the USB oral instructions, and the referee's of refigure place gesture track, finally reach the purpose of gesture identification.

Images