Finger motion recognition device, finger alphabet output method using said device, and glove-type device for operation in virtual space

Abstract

The objective of the present invention is to provide a finger motion recognition device and a finger alphabet output method using the device, which uses the device to recognize finger and hand movements such as opening, bending, and placing fingers together with a high degree of accuracy, converts the movements into finger alphabets, and outputs the results as voice, text, Braille, or other means via finger alphabets on an external device such as a personal computer or smartphone, thereby enabling smooth communication between visually impaired people and people in another space. [Solution] The present invention provides a hand movement recognition device 1 having a movement detection unit 2 that uses a sensor to detect multiple predetermined types of human movements, and a finger alphabet information output unit 3 that outputs finger alphabet information for inputting finger alphabets corresponding to human movements to an external device UT, where the movement detection unit 2 is worn on the five fingers of the hand and has a glove-type device 2 equipped with sensors that detect the movements of each of the five fingers.

Description

[Technical field]

[0001] The present invention relates to a hand and finger gesture recognition device that recognizes hand and finger gestures, a finger alphabet output method using the device, and a glove-type device that allows an operator to operate his / her avatar in a virtual space (such as the metaverse). In particular, the present invention relates to a technology that detects the shape and orientation of hands and fingers using a glove-type device and recognizes this as finger alphabet. [Background technology]

[0002] Conventionally, there are known techniques for detecting finger movements and using them as input information. For example, Patent Document 1 discloses a technique for detecting finger movements using a glove-type device. Patent Document 2 discloses a technique for recognizing a specific movement based on finger movements. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 6-138815 [Patent Document 2] JP 2014-38543 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, communication between visually impaired people and able-bodied people is based on Braille or voice output, and although technology that recognizes the shape and movement of hands and fingers is being researched, in order for everyone to use these communication tools, they need to use a device specifically for Braille or learn voice output technology. In particular, if the Braille or finger spelling is output using a device and it is different from the intended Braille or finger spelling, it is cumbersome to cancel the Braille or finger spelling and re-input the correct Braille or finger spelling.

[0005] Therefore, the object of the present invention is to provide a finger movement recognition device and a finger alphabet output method using the device, which can convert finger and hand movements such as opening, bending, and placing fingers together into finger alphabets by using the device, and also enables additional operations such as confirming or canceling the converted finger alphabet, and further enables the conversion result to be output as voice via finger alphabets on an external device such as a personal computer or smartphone, thereby facilitating smooth communication between visually impaired people and able-bodied people. [Means for solving the problem]

[0006] In order to achieve this object, the present invention has a motion detection unit that detects a plurality of predetermined types of human motions by a sensor, and a finger alphabet information output unit that outputs finger alphabet information for inputting finger alphabets corresponding to the human motions to an external device, and the motion detection unit has a glove-type device that is worn on five fingers of a hand and detects the motions of the five fingers to recognize the finger alphabet and then confirms or cancels the finger alphabet. The motion detection unit further includes a leg-mounted device that detects a predetermined foot motion using an acceleration sensor attached to the foot to determine or cancel the finger alphabet, and the finger alphabet information output unit outputs the finger alphabet information to an external device when the foot motion is detected. This is a hand and finger gesture recognition device.

[0007] In addition, the sensor of the glove-type device may include a first to fifth sensor group attached to the joint positions on the back of each of the five fingers, a sixth sensor attached to either the palm or the back of the hand, and a seventh sensor attached to the wrist, where the first to fifth sensor groups are bending sensors that detect the bending and straightening of the fingers, the sixth sensor is a sensor that detects the inclination of the hand, and the seventh sensor is a sensor that detects the direction of movement of the hand. Effect of the Invention

[0012] The present invention detects finger and hand movements, recognizes them as finger alphabets, converts these movements into finger alphabets, and also enables additional operations such as confirming or canceling the converted finger alphabet.Furthermore, by connecting to an external device, it is possible to output audio via finger alphabets, thereby achieving the effect of smooth communication between visually impaired people and people in another space. [Brief description of the drawings]

[0013] [Figure 1] 1 is a schematic configuration diagram of a hand movement recognition device 1 according to an embodiment of the present invention. [Diagram 2] FIG. 1 is a front view of a glove-type device 20 according to a first embodiment. [Diagram 3] FIG. 2 is a rear view of the glove type device 20 of the first embodiment. [Figure 4] FIG. 2 is a left side view of the glove type device 20 of the first embodiment. [Diagram 5] FIG. 11 is a perspective view of a leg-mounted device 28 according to a second embodiment. [Figure 6] FIG. 11 is a perspective view of a leg-mounted device 29 according to a second embodiment. [Figure 7] 2 is a flowchart of a finger character output method according to a first embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings. In the following, the same elements in all the drawings are given the same reference numerals, and duplicated explanations will be omitted. In addition, in the explanation in the text, the reference numerals previously described will be used as necessary.

[0015] FIG. 1 shows a schematic configuration diagram of a hand gesture recognition device 1 according to an embodiment of the present invention. As shown in FIG. 1, the hand gesture recognition device 1 according to the present invention is roughly divided into a gesture detection unit 2 for detecting gestures of a person (operator) and a finger character information output unit 3 for outputting information as finger characters corresponding to the gestures detected by the gesture detection unit 2 to an external device UT, and these are connected to each other via a wiring H1. Furthermore, the finger character information output unit 3 can output as voice via finger character data by connecting to an external device UT such as a smartphone, a personal computer, or a tablet terminal via wirings H2 and H3. Furthermore, the finger character information output unit 3 can output information as finger characters as braille by connecting to a braille output device such as a braille printer or a braille display. The gesture detection unit 2 and the finger character information output unit 3 will be described in detail below.

[0016] <Motion detection unit 2> The motion detection unit 2 is an element constituting the hand motion recognition device of the present invention, which detects multiple predetermined types of motions of a person's hands, feet, etc., using sensors. When the motion detection unit 2 is a glove-type device as shown in Fig. 1, it detects the motions of the five fingers of a person's (operator's) hand and collects electrical signals sent from various sensors provided in the glove-type device as data.

[0017] The electrical signals collected as data are transmitted to a finger alphabet information output unit described later via wireless or wired communication means. In addition to the glove-type device described above, the motion detection unit 2 can also use a leg-mounted device that detects a predetermined foot motion using a sensor attached to the thigh or ankle. Details of the glove-type device 20, which is a first embodiment of the motion detection unit 2, and the leg-mounted devices 28 and 29, which are second embodiments, will be described below.

[0018] (Glove-type device 20) FIG. 2 shows a front view of a glove-type device 20 as a first embodiment of the motion detection unit, FIG. 3 shows a rear view of the glove-type device 20, and FIG. 4 shows a left side view of the glove-type device 20. The glove-type device 20 as a first embodiment of the motion detection unit is a device equipped with a function of detecting the motion of five fingers of a hand with high accuracy by attaching various sensors to a glove body made of a durable material such as fiber-reinforced plastic or Kevlar fiber. By the action of the various sensors attached to the glove body, for example, the bending sensor measures the bending degree of the fingers, and the acceleration sensor captures the movement of the entire hand, it is possible to accurately capture minute movements such as bending and straightening of the fingers and the movement of the wrist as information of an electrical signal. Here, the bending sensor is, for example, two thin elastic plates each having a resistance such as a strain gauge formed on the surface and the like, bonded together. The degree of bending of the elastic body can be detected by detecting the dimensional change of the strain gauge generated on the front side and the back side as a resistance value when the elastic body is bent. In addition, an acceleration sensor has a weight and a sensor element inside, and when the weight moves, a compressive or tensile force is generated in the sensor element. The change in resonance frequency caused by this compressive or tensile force can be detected as vibration, inclination, or linear motion generated in an object using a predetermined calculation formula.

[0019] The glove-type device 20 is a glove-type device in which a plurality of sensors are distributed and attached as shown in Figures 2 to 4, and can be used by inserting a hand inside. The sensors attached to the glove-type device 20 are formed of at least first to fifth sensor groups 21 to 25 attached to the joint positions on the back side of each of the five fingers (thumb, index finger, middle finger, ring finger, and little finger), a sixth sensor 26 attached to the palm or back of the hand, and a seventh sensor 27 attached to the wrist, as shown in Figures 2 to 4.

[0020] 2, the first sensor group 21 is attached to the thumb, the second sensor group 22 to the index finger, the third sensor group 23 to the middle finger, the fourth sensor group 24 to the ring finger, and the fifth sensor group 25 to the little finger. Furthermore, the sensors attached to each finger are divided into three types according to the joint position of each finger.

[0021] For example, the first sensor group 21 attached to the thumb is divided into three types of sensors: a first upper end sensor 21A attached to the first joint (the joint at the most distal end) of the thumb, a first central sensor 21B attached to the second joint of the thumb, and a first lower end sensor 21C attached to the third joint of the thumb, as shown in Fig. 2. The first upper end sensor 21A, the first central sensor 21B, and the first lower end sensor 21C are connected to each other by a wiring W21 as shown in Fig. 2, and the wiring W21 is connected to a wiring H1 connected to the finger character information output unit 3 described later.

[0022] Therefore, the degree of bending of the thumb is detected by each sensor and transmitted as an electric signal to a finger character information output unit 3, which is connected via a wiring H1 shown in Fig. 2. The second to fifth sensor groups 22 to 25 attached to the other four fingers, the index finger, the middle finger, the ring finger, and the little finger, have the same configuration, and therefore detailed description thereof will be omitted.

[0023] The sensors applied to the first to fifth sensor groups 21 to 25 are preferably bend sensors that can detect the degree of bending of the fingers (how much each finger is bent) by arranging them at the dorsal (nail side) joints of each finger. As shown in FIG. 2, the first sensor group 21 attached to the thumb is illustrated as having all three types of sensors, the first upper end sensor 21A, the first central sensor 21B, and the first lower end sensor 21C, attached to the dorsal (nail side). However, by attaching the sensors to the side or ventral side (palm side) of the thumb, the strain gauge in the attached bend sensor changes in deformation, changing the resistance value, and the degree of bending of the thumb, that is, the degree of bending of the thumb, such as opening the thumb outward or placing it on the index finger, can be detected. Furthermore, by attaching a bend sensor to the joint position of the finger, for example, when the index finger (index finger) is bent 90 degrees, it can be converted into a specific finger alphabet by setting it in advance as a gesture that represents a specific finger alphabet.

[0024] The sixth sensor 26 is a sensor for detecting the state of the hand (part from the wrist to the tip), such as tilt or twist, by being attached to the palm of the hand as shown in Figures 3 and 4. The sixth sensor 26 is connected to a seventh sensor 27 (described later) via a wiring W26.

[0025] In addition, by pre-setting, for example, a state in which only the hand is moved without bending the fingers as a gesture indicating confirmation or cancellation of a specific input content, the sixth sensor 26 can perform actions such as tilting the hand by tilting the part from the wrist forward or backward, or twisting the part from the wrist to the left or right as a function for inputting or outputting finger alphabets.

[0026] Therefore, it is preferable that the type of sensor is a bending sensor or an acceleration sensor. Although the sixth sensor shown in Fig. 3 and Fig. 4 is attached to the palm of the hand, it may be attached to the back of the hand.

[0027] The seventh sensor 27 is a sensor that is attached to the wrist as shown in Figs. 3 and 4 and detects the action of tilting the hand (or forearm) to the front (operation side) or the back (opponent side) or the action of bending the hand (or forearm) to the left or right. The seventh sensor 27 is connected to the sixth sensor 26 described above via a wiring W26, and is also connected to a wiring H1 that is connected to the finger alphabet information output unit via a separate wiring W27. The seventh sensor is preferably an acceleration sensor or a vibration sensor. Here, the vibration sensor is, for example, a sensor that has a piezoelectric element sandwiched between two weights inside. When an object vibrates, the piezoelectric element deforms in response to the movement of the two weights and outputs an electric signal, and the vibration of the object can be detected from the measured voltage.

[0028] As described above, the glove-type device is equipped with the various sensors described above, and can capture the bending and straightening of each of the five fingers and the up / down / left / right movements of the wrist as electrical signal information. Therefore, for example, based on small operations such as opening, placing, and bending the thumb (upper finger) as described above, a series of steps such as confirming or canceling characters in addition to displaying letters, numbers, punctuation marks, etc. corresponding to the 50 Japanese syllabary can be handled by a single glove-type device alone.

[0029] (Leg-mounted devices 28, 29) Fig. 5 shows a perspective view of a leg-mounted device 28 of a second embodiment of the motion detection unit wrapped around the thigh and fixed thereto, and Fig. 6 shows a perspective view of a leg-mounted device 29 of the second embodiment of the motion detection unit wrapped around the ankle. The leg-mounted device of the second embodiment of the motion detection unit is equipped with various sensors such as an acceleration sensor, and is a device that is fixed to a person's leg when used. For example, there is a type (leg-mounted device 28) in which a belt-type part is wrapped around the thigh and fixed thereto as shown in Fig. 5, and a type (leg-mounted device 29) in which a ring-shaped part is attached to the ankle as shown in Fig. 6.

[0030] Moreover, the leg-mounted device 28 has a wiring W28 as shown in FIG. 5, which is connected to the wiring H1 connected to the finger alphabet information output unit 3. Similarly, the leg-mounted device 29 has a wiring W29 as shown in FIG. 6, which is connected to the wiring H1 connected to the finger alphabet information output unit. Furthermore, the leg-mounted devices 28, 29 are devices that have the function of accurately detecting movements such as swinging up (or down) the leg or kicking up by attaching them to the thighs or ankles as shown in FIG. 5 and FIG. 6. By using the functions of various sensors attached to these devices, for example, an acceleration sensor can grasp the instantaneous movement of the thighs in the up and down directions, and accurately capture such movements as electrical signal information.

[0031] As described above, by using the leg-mounted device in addition to the glove-type device 20, for example, steps such as confirming or canceling the finger alphabet displayed using the glove-type device can be individually handled by swinging up the leg wearing the leg-mounted device. Therefore, while performing a series of steps from displaying the finger alphabet to confirming it with one hand and foot (for example, the right hand and right foot), it is also possible to use the other hand (for example, the left hand) to operate a device such as sending an email using an external device such as a smartphone, a personal computer, or a tablet terminal.

[0032] <Finger alphabet information output unit 3> The finger alphabet information output unit 3 is a device that generates finger alphabet information corresponding to the human movements detected by the above-mentioned movement detection unit and outputs it to an external device. It analyzes the data sent from the movement detection unit, generates finger alphabet information based on the correspondence between the movements and finger alphabets that have been determined in advance, and transmits it to the external device.

[0033] The finger alphabet information output unit 3 analyzes the motion data, maps the motions and finger alphabets, generates finger alphabet information, and transmits it to an external device as finger alphabet data via wired or wireless means. Note that the finger alphabet information output unit 3 is illustrated as a device separate and independent from the motion detection unit 2 as shown in FIG. 1, but it can also be integrated by being embedded in a device such as the glove-type device or leg-mounted device of the motion detection unit 2.

[0034] Next, an embodiment of a method for outputting finger characters using a hand gesture recognition device will be described. The method for outputting finger characters using a hand gesture recognition device of the present invention is mainly composed of the following first to ninth steps. First step: Sensor detection based on hand movement 2nd step: Transmitting an electrical signal when motion is detected 3rd step: Check whether electrical signals are received -Fourth step: Selecting the corresponding handwriting based on the analysis of the electrical signals - 5th step: Confirmation of finger alphabet selection by analyzing electrical signals - 6th step: Audio output of the selected finger alphabet on an external device -7th step: Check the selected handwriting Step 8: Confirm the selected handwriting 9th step: Output of finger spelling data to external devices using voice, text, and Braille The following describes the details of steps 1 to 9 when the glove-type device shown in Figures 2 to 4 or the leg-mounted device shown in Figures 5 and 6 is used as the motion detection unit, with reference to the drawings. Figure 7 shows a flowchart of a method for outputting finger alphabet characters using a hand motion recognition device according to one embodiment of the present invention.

[0035] (1st step: Sensor detection based on hand and finger movements) First, a person (operator) wears a glove-type device on one hand. This glove-type device is equipped with multiple types of sensors, such as a bending sensor and an acceleration sensor, and detects the movements of the hand and five fingers (first step S001 shown in Fig. 7).

[0036] The bending sensor accurately detects the bending direction and bending angle of each finger joint and measures the degree of bending of the finger by capturing changes in the resistance value of the strain gauge inside the sensor according to the finger movement. In addition, the acceleration sensor detects the direction and speed of hand movement and determines the speed and direction of movement by measuring the acceleration associated with the hand movement.

[0037] The output of these sensors can be continuously monitored and calibrated as necessary to maintain measurement accuracy. For example, recalibration during initial use or at regular intervals can correct any misalignment or errors in the sensors, ensuring that accurate data is always obtained.

[0038] (2nd step: Transmitting an electrical signal when motion is detected) In a second step, the electric signal detected in the first step described above is transmitted from the glove type device 20 to the finger character information output unit 3 (second step S002 shown in FIG. 7). The glove type device 20 can transmit the electric signal to the finger character information output unit quickly and reliably via a wired or wireless communication means.

[0039] Specifically, by using Bluetooth (registered trademark) or Wi-Fi (registered trademark) for wireless communication, it is possible to transmit information quickly without being subject to the physical constraints of wired communication such as cables. In addition, by using a recording medium such as a USB for wired communication, high-speed and stable communication is possible.

[0040] (Third step: Check whether electrical signals are received or not) The finger character information output unit 3 checks whether an electric signal has been received from the glove type device 20. If the finger character information output unit 3 has not received an electric signal, it sends a command to the glove type device 20 to retransmit the electric signal (retransmission command: third step S003).

[0041] At this time, by transmitting the time stamp of the received signal and the identification information of the sensor together, it becomes possible to identify the time of generation of the signal and the source of the signal. In this way, the signal transmitted from the glove type device 20 is appropriately managed and processed by the finger character information output unit 3, so that the signal can be smoothly transferred to the next step.

[0042] In order to increase the reliability of communication, a security protocol such as encryption technology can be introduced into the communication between the glove type device 20 and the finger character information output unit 3. This prevents a third party from intercepting or tampering with the electric signal, ensuring highly confidential communication. In this way, in the second step, an electric signal is transmitted from the glove type device to the finger character information output unit 3, and it is ensured that the signal is received accurately and safely.

[0043] (4th step: Selection of corresponding finger spelling based on analysis of electrical signals) (5th step: Confirm finger alphabet selection by analyzing electrical signals) In the fourth and fifth steps, the finger alphabet information output unit 3 analyzes the electrical signal received in the second step (fourth step S004 in FIG. 7) to select a finger alphabet corresponding to the electrical signal (fifth step S005 in FIG. 7). The finger alphabet information output unit 3 includes a plurality of processes for analyzing the received electrical signal. This analysis process is composed of a signal pre-processing process, a feature extraction process, and a finger alphabet correspondence process.

[0044] First, the signal pre-processing process involves noise removal and normalization of the received electrical signal. For noise removal, filtering techniques are used to remove noise from the external environment and the device itself. For example, applying a low-pass filter can remove high-frequency and low-frequency components from the electrical signal to improve the quality of the electrical signal. Normalization is performed to unify the range and scale of the electrical signal obtained from the sensor, and is an important process for improving the accuracy of analysis.

[0045] Next, in a feature extraction process, important features are extracted from the preprocessed electrical signal. The extracted features include the bending angle of each finger, the speed and direction of the hand movement, and the pressure on the fingertips. These features are analyzed using a machine learning algorithm or an existing finger alphabet recognition technology. For example, an algorithm such as a neural network can be used to identify the corresponding finger alphabet from the extracted features with high accuracy. Note that, if the corresponding finger alphabet cannot be selected in the fifth step S005 as a result of the analysis in the fourth step S004 shown in FIG. 7, a command to reanalyze the electrical signal is sent to the finger alphabet information output unit 3 (reanalysis command).

[0046] (Sixth step: Voice output of selected finger alphabet on external device) Finally, in the finger alphabet matching process, the extracted features are used to select the corresponding finger alphabet using a pre-trained model, which is pre-trained to match each specific finger flexion pattern or hand movement to a specific finger alphabet (finger alphabet).

[0047] For example, a highly accurate model is constructed based on a large amount of data using machine learning technology. This allows the finger alphabet corresponding to the operator's hand and finger movements to be selected quickly and accurately. The selected finger alphabet is then displayed on the screen of an external device such as a personal computer or smartphone connected to the finger alphabet information output unit via wired or wireless communication means, and is simultaneously output as voice using an application (app) with a reading function and a speaker of the external device (sixth step S006 shown in FIG. 7).

[0048] (7th step: Confirm the selected finger alphabet) In the seventh step, the operator performs a confirmation action to finally determine the finger alphabet selected in the fifth step (seventh step S007 shown in FIG. 7). Specifically, the finger alphabet selected in the fifth step is displayed on an external device connected to the finger alphabet information output unit 3, and is simultaneously output as voice using an application (app) and a speaker with a reading function of the external device. If the voice-output finger alphabet matches the finger alphabet intended by the operator, the operator confirms it by performing a specific action using the hand or finger. This confirmation action can be a decision to determine the selected finger alphabet by a predetermined action pattern, such as an action of tilting the hand forward, an action of raising the hand (arm), or an action of moving the hand (arm) outside the body.

[0049] If the selected digit alphabet is different from the digit alphabet intended by the operator, an action for canceling the digit alphabet can be performed (seventh step S007 shown in FIG. 7). For example, when the glove-type device 20 is used as the action detection unit, an electrical signal for canceling the selected digit alphabet can be transmitted by performing a specific action such as bending the hand toward the operator, lowering the hand (arm), or moving the hand (arm) inwardly.

[0050] Furthermore, when a leg-mounted device is used as the motion detection unit 2, an electrical signal can be sent to cancel the selected finger alphabet by performing a motion of lifting the thigh or kicking a ball or the like. This prevents an incorrect finger alphabet from being determined, and enables the finger alphabet to be input again accurately. After re-inputting the finger alphabet, a command is sent to re-display the re-inputted finger alphabet on the screen of an external device and output it as voice at the same time (re-display command).

[0051] (8th step: Confirmation of selected finger spelling) When the operator performs a confirmation action, the glove type device again transmits an electrical signal corresponding to that action to the finger alphabet information output unit (eighth step S008 shown in FIG. 7). This electrical signal is for finalizing the previously selected finger alphabet, and indicates that the confirmation action has been performed properly. At this time, in order to correctly detect the confirmation action, the glove type device detects the action with high sensitivity, and a filtering process is performed to prevent erroneous recognition.

[0052] If the confirmation operation is performed correctly and the selected finger alphabet is finally confirmed, the finger alphabet is prepared to be output to an external device in the next step 5. In this way, in the fourth step, the process in which the selected finger alphabet is finally confirmed through the confirmation operation is managed in detail.

[0053] (9th step: Voice output of finger spelling data to external device using voice, characters, and Braille) In a ninth step, the finger alphabet data finally determined in the eighth step is displayed as characters from the finger alphabet information output unit to an external device, and is also output as any of the following means: voice, characters, braille, etc. (ninth step S009 shown in FIG. 7). This output step is performed between the finger alphabet information output unit and the external device via a wired or wireless communication means. The external device includes a wide variety of devices such as computers, smartphones, tablets, and various types of Internet-connected devices.

[0054] Bluetooth (registered trademark) and Wi-Fi (registered trademark) are commonly used as wireless communication means. Bluetooth (registered trademark) is suitable for short-distance data transmission with low power consumption, and provides stable communication between paired devices. On the other hand, Wi-Fi enables high-speed data transfer over a wide area, and also enables communication via the Internet using a home network or public wireless LAN. USB and serial cables are used as wired communication means, and provide high-speed and stable data transfer.

[0055] When sending finger alphabet data, an error check function and encryption technology are applied to ensure the reliability and integrity of the transmitted data. For example, by adding a checksum to the transmitted data, the receiving side can detect data errors and request retransmission if necessary. In addition, the confidentiality of the transmitted data and the security of communication can be ensured by using encryption technology.

[0056] The transmitted finger alphabet data is used by various applications and software on the external device. For example, it can be used as finger alphabet input in real time in text editors, messaging applications, email clients, etc. This also enables quick and efficient information exchange between different users, and realizes an intuitive finger alphabet input interface through hand movements.

[0057] In addition, the process of sending finger alphabet data can be customized according to the user's needs, allowing for a variety of usage scenarios, such as a mode for direct input to a specific application, or a mode for sending data to a cloud service and synchronizing it with remote devices.

[0058] Thus, in the final ninth step, the finger alphabet data finally confirmed in the eighth step is transmitted to an external device via wireless or wired communication means, and the processes used in various applications and software are managed in detail.

[0059] By using the above-mentioned hand gesture recognition device, any kind of finger alphabet can be expressed, and further, by using the finger alphabet output method using the device, the device can accurately detect the movements of the fingers and hands, recognize them as finger alphabets, convert those movements into finger alphabets, and perform additional operations such as deciding or confirming the converted finger alphabet. By connecting to an external device, it is possible to output the finger alphabet data as voice. Therefore, it is possible to smoothly communicate between visually impaired people and people in different spaces. [Explanation of symbols]

[0060] 1. Hand movement recognition device 2. Motion detection unit 3 Finger alphabet information output section 20 Glove-type device 21~25 1st~5th sensor group 21A~21C 1st (top, center, bottom) sensor group 22A~22C 2nd (top, center, bottom) sensor group 23A~23C 3rd (top, center, bottom) sensor group 24A~24C 4th (top, center, bottom) sensor group 25A~25C 5th (top, center, bottom) sensor group 26 6th Sensor 27 7th Sensor 28,29 Leg-mounted devices H1~H3 wiring S001~S009 Steps 1~9 UT external equipment W21~W29 Wiring

Claims

1. a motion detection unit that detects a plurality of predetermined types of human motions using a sensor; a finger alphabet information output unit that outputs finger alphabet information for inputting a finger alphabet corresponding to the human motion to an external device; It has the motion detection unit has a glove-type device that is worn on five fingers of a hand and detects motions of the five fingers to recognize the finger alphabet and then confirms or cancels the finger alphabet, The finger movement recognition device is characterized in that the movement detection unit further has a leg-mounted device that determines or cancels the finger alphabet by detecting a predetermined foot movement using an acceleration sensor attached to the foot, and the finger alphabet information output unit outputs the finger alphabet information to an external device when the foot movement is detected.

2. 2. The hand movement recognition device according to claim 1, wherein the sensors of the glove-type device include first to fifth sensor groups attached to the joint positions on the back of each of the five fingers, a sixth sensor attached to either the palm or the back of the hand, and a seventh sensor attached to the wrist, wherein the first to fifth sensor groups are bending sensors that detect the bending and straightening of the fingers, the sixth sensor is a sensor that detects the inclination of the hand, and the seventh sensor is a sensor that detects the direction of movement of the hand.

Images