Method and apparatus for recognizing gestures

Abstract

The present invention relates to human-computer interaction technology, and in particular to a method for identifying gestures, a device for implementing the method, a wearable glove comprising the device, and a computer storage medium for implementing the method. According to one aspect of the present invention, the method for identifying gestures comprises the following steps: generating a first vector associated with the bending angle of one or more finger joints and a second vector associated with the spatial orientation of the palm; determining the similarity between the first vector and the second vector and their respective reference vectors; and identifying the gesture according to the similarity, wherein the bending angle and the spatial orientation are obtained using a sensor.

Description

Technical Field

[0001] The present invention relates to human-computer interaction technology, and in particular to a method for identifying gestures, a device and system for implementing the method, a wearable device including the device, and a computer storage medium for implementing the method. Background Art

[0002] Gesture recognition is an important aspect of human-computer interaction. Its purpose is to enable users to control or interact with devices using simple gestures, thereby establishing a richer and simpler communication method between machines and humans than text user interfaces and graphical user interfaces.

[0003] In the prior art, whether the gesture is static or dynamic, its recognition process usually includes the following steps: gesture image acquisition, gesture detection and segmentation, gesture analysis and gesture recognition. Gesture segmentation is a key step in the recognition process, and its effect directly affects the execution effect of subsequent steps. Currently, the commonly used gesture segmentation methods mainly include gesture segmentation based on monocular vision and gesture segmentation based on stereo vision. The former uses an image acquisition device to obtain a gesture and obtain a planar model of the gesture, while the latter uses multiple image acquisition devices to obtain different images of the gesture and convert them into a three-dimensional model. Gesture recognition is the process of classifying the trajectories in the model parameter space into a subset of the space. Common recognition algorithms include template matching neural network method and hidden Markov model method.

[0004] In real applications, gesture recognition is often interfered by environmental factors and causes misrecognition (for example, too bright or too dark light and small difference between gesture and background may cause inaccurate gesture segmentation). In addition, the above recognition algorithm requires a large amount of data to train the model, which is time-consuming and laborious. Therefore, it is urgent to provide a gesture recognition method and device that can overcome the above shortcomings of the prior art. Summary of the invention

[0005] An object of the present invention is to provide a method for recognizing gestures, which has the advantages of being easy to implement and having a high accuracy rate.

[0006] A method for identifying a gesture according to one aspect of the present invention comprises the following steps:

[0007] generating a first vector associated with a bending angle of one or more finger joints and a second vector associated with a spatial orientation of a palm;

[0008] determining similarities of the first and second vectors to respective reference vectors; and

[0009] recognizing the gesture according to the similarity,

[0010] The bending angle and spatial orientation are obtained by using sensors.

[0011] Preferably, in the above method, the bending angle is obtained by setting a sensor on an area of ​​the wearable device corresponding to the phalanges near the finger joints.

[0012] Preferably, in the above method, the spatial orientation is obtained by arranging a sensor on an area of ​​the wearable device corresponding to the back of the hand or the palm of the hand.

[0013] Preferably, in the above method, the second vector is in the form of a quaternion.

[0014] Preferably, in the above method, the similarity between the first vector and the corresponding reference vector is measured by the Euclidean distance between the two, and the similarity between the second vector and the corresponding reference vector is measured by the angle between the two.

[0015] According to another aspect of the present invention, a method for identifying a gesture comprises the following steps:

[0016] generating a first vector associated with a bending angle of one or more finger joints;

[0017] Determining the similarity between the first vector and a reference vector; and

[0018] recognizing the gesture according to the similarity,

[0019] The bending angle is obtained by using a sensor.

[0020] Another object of the present invention is to provide a device for identifying gestures, which has the advantages of simple implementation and high accuracy.

[0021] According to another aspect of the invention, the device comprises:

[0022] A first module for generating a first vector associated with a bending angle of one or more finger joints and a second vector associated with a spatial orientation of a palm;

[0023] A second module is used to determine the similarity between the first vector and the second vector and their respective reference vectors; and

[0024] The third module is used to recognize the gesture according to the similarity,

[0025] The bending angle and spatial orientation are obtained by using sensors.

[0026] According to another aspect of the invention, the device comprises:

[0027] A first module, for generating a first vector associated with a bending angle of one or more finger joints;

[0028] A second module is used to determine the similarity between the first vector and a reference vector; and

[0029] The third module is used to recognize the gesture according to the similarity,

[0030] The bending angle is obtained by using a sensor.

[0031] According to another aspect of the present invention, an apparatus comprises a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the program is executed to implement the method as described above.

[0032] Another object of the present invention is to provide a wearable device which has the advantages of simple implementation and high accuracy when recognizing gestures.

[0033] A wearable device according to another aspect of the present invention comprises:

[0034] A first sensor is disposed on the wearable device in an area corresponding to a phalanx near a finger joint to obtain a bending angle of the finger joint;

[0035] A second sensor is disposed on the wearable device in an area corresponding to the back of the hand or the palm to obtain the spatial orientation of the palm;

[0036] The device for recognizing gestures comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the following steps are implemented by executing the program:

[0037] generating a first vector associated with a bending angle of one or more finger joints and a second vector associated with a spatial orientation of a palm;

[0038] determining similarities of the first and second vectors to respective reference vectors; and

[0039] The gesture is recognized based on the similarity.

[0040] A wearable device according to another aspect of the present invention comprises:

[0041] A first sensor is disposed on the wearable device in an area corresponding to a phalanx near a finger joint to obtain a bending angle of the finger joint;

[0042] The device for recognizing gestures comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the following steps are implemented by executing the program:

[0043] generating a first vector associated with a bending angle of one or more finger joints;

[0044] determining a similarity between the first vector and a respective reference vector; and

[0045] The gesture is recognized based on the similarity.

[0046] Another object of the present invention is to provide a system for recognizing gestures, which has the advantages of simple implementation and high accuracy.

[0047] A system for recognizing gestures according to another aspect of the present invention comprises:

[0048] Wearable devices, including:

[0049] A first sensor is disposed on the wearable device in an area corresponding to a phalanx near a finger joint to obtain a bending angle of the finger joint;

[0050] A second sensor is disposed on the wearable device in an area corresponding to the back of the hand or the palm to obtain the spatial orientation of the palm;

[0051] A computing device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the following steps are implemented by executing the program:

[0052] generating a first vector associated with a bending angle of one or more finger joints and a second vector associated with a spatial orientation of a palm;

[0053] determining similarities of the first and second vectors to respective reference vectors; and

[0054] The gesture is recognized based on the similarity.

[0055] According to another aspect of the present invention, a system for recognizing gestures comprises:

[0056] Wearable devices, including:

[0057] A first sensor is disposed on the wearable device in an area corresponding to a phalanx near a finger joint to obtain a bending angle of the finger joint;

[0058] A computing device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the following steps are implemented by executing the program:

[0059] generating a first vector associated with a bending angle of one or more finger joints;

[0060] determining a similarity between the first vector and a respective reference vector; and

[0061] The gesture is recognized based on the similarity.

[0062] The present invention also provides a computer-readable storage medium storing a computer program, wherein the computer program implements the method described above when executed by a processor.

[0063] In the present invention, the bending angle of the finger joints and the spatial orientation of the palm are used to characterize the gesture features. Since the bending angle and the spatial orientation are both measured by sensors, compared with the prior art, the recognition accuracy is improved and the complexity caused by the use of image recognition algorithms is avoided. In addition, since the sensor is very suitable for being set on a wearable device, the device for identifying gestures of the present invention can be well integrated into the wearable device. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The above and / or other aspects and advantages of the present invention will become clearer and easier to understand through the following description in conjunction with the accompanying drawings, in which the same or similar elements are represented by the same reference numerals. The accompanying drawings include:

[0065] Figure 1 is a schematic diagram, which exemplarily shows the distribution of multiple sensors on a wearable device.

[0066] Figure 2 FIG. 4 is a schematic diagram of a method for identifying gestures according to an embodiment of the present invention.

[0067] Figure 3 FIG. 4 is a schematic diagram of a method for recognizing gestures according to another embodiment of the present invention.

[0068] Figure 4 FIG. 4 is a schematic block diagram of a device for identifying gestures according to another embodiment of the present invention.

[0069] Figure 5 The present invention is a schematic block diagram of a device for recognizing gestures according to another embodiment of the present invention.

[0070] Figure 6 The present invention is a schematic block diagram of a device for recognizing gestures according to another embodiment of the present invention.

[0071] Figure 7 The present invention is a schematic block diagram of a wearable device according to another embodiment of the present invention.

[0072] Figure 8 The present invention is a schematic block diagram of a wearable device according to another embodiment of the present invention.

[0073] Fig. 9 The present invention is a schematic block diagram of a system for recognizing gestures according to another embodiment of the present invention.

[0074] Fig.10 The present invention is a schematic block diagram of a system for recognizing gestures according to another embodiment of the present invention. DETAILED DESCRIPTION

[0075] The present invention is described more fully below with reference to the accompanying drawings in which exemplary embodiments of the present invention are illustrated. However, the present invention may be implemented in different forms and should not be construed as being limited to the embodiments given herein. The above embodiments are given to make the disclosure herein comprehensive and complete, so as to more fully convey the scope of the present invention to those skilled in the art.

[0076] In this specification, terms such as "comprise" and "include" indicate that in addition to the units and steps directly and explicitly stated in the specification and claims, the technical solution of the present invention does not exclude the situation where it has other units and steps that are not directly or explicitly stated.

[0077] In this specification, terms such as "first" and "second" do not indicate the order of units in terms of time, space, size, etc., but are only used to distinguish between units.

[0078] In this specification, “coupling” should be understood to include a situation where electric energy or electric signals are directly transmitted between two units, or a situation where electric energy or electric signals are indirectly transmitted via one or more third units.

[0079] In this specification, the "bending angle" of the finger joint refers to the relative angle of the phalanges near the finger joint, and the "spatial orientation" of the palm refers to the normal direction of the plane where the palm is located.

[0080] According to one aspect of the present invention, the bending angle of one or more finger joints is used to characterize the gesture feature, that is, the bending angle is used as a characteristic parameter for gesture recognition. In order to improve the recognition accuracy and meet the needs of diversified gesture types, the combination of the bending angle of the finger joints and the spatial orientation of the palm can also be used to characterize the gesture feature, that is, the bending angle and the spatial orientation are used as characteristic parameters for gesture recognition at the same time.

[0081] According to another aspect of the present invention, a sensor is used to measure the bending angle of the finger joints and the spatial orientation of the palm. Preferably, the sensor can be set on the wearable device in the area corresponding to the phalanges near the finger joints to measure the Euler angles of the phalanges, thereby determining the angle between the phalanges or the bending angle of the joints. In addition, preferably, the sensor can be set on the wearable device in the area corresponding to the back of the hand or the palm to measure the spatial orientation of the palm.

[0082] Figure 1is a schematic diagram, which exemplarily shows the distribution of multiple sensors on a wearable device, where the numbers represent the numbers of the sensors. Figure 1 It can be seen that sensors 1-11 are arranged on the phalanges on both sides of the finger joints, and sensor 12 is arranged on the back of the hand.

[0083] Various sensors can be used to measure the bending angle and spatial orientation. In one embodiment of the present invention, a MEMS sensor can be used, which can measure the acceleration of the object along the three coordinate axes of the rectangular coordinate system, the angular velocity of the object rotating around the three coordinate axes, and the magnetic field along the three coordinate axes.

[0084] Figure 2 FIG. 4 is a schematic diagram of a method for identifying gestures according to an embodiment of the present invention.

[0085] like Figure 2 As shown, in step 210, the apparatus for identifying gestures receives the Euler angles of the phalanges of the fingers and the Euler angles of the back of the hand measured by the sensors installed on the wearable device. As described above, the sensors can be arranged on the wearable device in the area corresponding to the phalanges near the finger joints and in the area corresponding to the back of the hand or the palm.

[0086] Then, the process proceeds to step 220 , where the apparatus for identifying gestures generates a first vector and a second vector based on the measurement values ​​of the sensor, wherein each element of the first vector represents a bending angle of one of the finger joints, and the second vector represents a spatial orientation of the palm.

[0087] Exemplarily, the first vector is denoted as J below. r ={J r (1),J r (2),…J r (i),…,J r (n)}, where i is the index of the finger joint, n is the number of finger joint feature parameters used to recognize gestures, and J r (i) represents the bending angle of the i-th finger joint or the angle between the phalanges on both sides of the joint, and the reference vector of the first vector corresponding to the k-th gesture type is recorded as J k ={J k (1),J k (2),…J k (i),…,J k (n)}, where J k (i) represents the i-th component of the reference vector.

[0088] In this embodiment, preferably, the second vector is in the form of a quaternion, that is, the second vector Q r= {Q1, Q2, Q3, Q4}. Although quaternions and Euler angles are mathematically equivalent when representing the spatial orientation of the palm, the use of quaternions is more conducive to the calculation of the spatial orientation transformation of the object, and can avoid the locking phenomenon that occurs when using Euler angles (i.e., the occurrence of singularities that cause calculation interruption).

[0089] Then, the process proceeds to step 230, where the apparatus for identifying gestures determines a first vector J r and the second vector Q r Similarity with respective reference vectors. It should be noted that for each gesture to be recognized, it has a corresponding reference vector for the first vector and the second vector, so the first vector and the second vector have multiple similarities, each similarity corresponding to a gesture type.

[0090] Preferably, for each gesture, a first vector J may be used. r The corresponding reference vector J k The Euclidean distance between them is used to measure their similarity, where k represents the type number of the gesture. Specifically, the similarity S with the kth gesture type can be calculated using the following formula: k1 :

[0091]

[0092] Here, J r (i) is the first vector J r The i-th element of k (i) is the reference vector J for the first vector of the k-th gesture type k The i-th element of , where n is the number of elements in the first vector.

[0093] Preferably, the second vector Q can be used r The corresponding reference vector Q k Specifically, the similarity S with the kth gesture type can be calculated using the following formula: k2 :

[0094]

[0095] Here, Q r (i) is the second vector Q r The i-th element of Q k (i) is the reference vector Q for the second vector k The ith element of .

[0096] Then, the process proceeds to step 240 , in which each similarity determined in step 230 is normalized to obtain a normalized similarity with a value range between 0 and 1.

[0097] Next, proceed to step 250 to identify the gesture based on the normalized similarity.

[0098] In this embodiment, the value of the normalized similarity can be comprehensively considered to identify the gesture. Taking the normalized similarities S' k1 and S' k2 of the first vector and the second vector with respect to the k-th gesture type as an example, the following rules can be used to determine the similarities of the first and second vectors with their respective reference vectors:

[0099] If S'1 < 0.05, it is determined that the first vector is highly similar to the reference vector; if 0.05 ≤ S'1 < 0.1, it is determined that the first vector is moderately similar to the reference vector; if 0.1 ≤ S'1 < 0.15, it is determined that the first vector is slightly similar to the reference vector; and if S'1 ≥ 0.15, it is determined that the first vector is not similar to the reference vector. For the normalized similarity S'2, similar determination rules can also be used, that is, if S'2 > 0.95, it is determined that the second vector is highly similar to the reference vector; if 0.90 < S'2 ≤ 0.95, it is determined that the second vector is moderately similar to the reference vector; if 0.85 < S'2 ≤ 0.90, it is determined that the second vector is slightly similar to the reference vector; and if S'2 ≤ 0.85, it is determined that the second vector is not similar to the reference vector.

[0100] After obtaining the judgment results for the first and second vectors according to the above rules, the current detected gesture can be further determined whether it is the k-th gesture type based on the judgment results. The following Table 1 exemplarily gives an example of determining the similarity degree between the current detected gesture and the k-th gesture type according to the results of the first and second vectors.

[0101] Table 1

[0102] Similarity <![CDATA[S′1<0.05]]> <![CDATA[0.05≤S′1<0.1]]> <![CDATA[0.1≤S′1<0.15]]> <![CDATA[S′1≥0.15]]> <![CDATA[S′2>0.95]]> Highly Similar Moderately similar Low similarity Not similar <![CDATA[0.90<S′2≤0.95]]> Moderately similar Moderately similar Low similarity Not similar <![CDATA[0.85<S′2≤0.90]]> Low similarity Low similarity Not similar Not similar <![CDATA[S′2≤0.85]]> Not similar Not similar Not similar Not similar

[0103] Figure 3 It is a schematic diagram of a method for identifying gestures according to another embodiment of the present invention.

[0104] Compared with Figure 2 the embodiment shown, this embodiment only uses the bending angle of the finger as the characteristic variable for gesture recognition.

[0105] As Figure 3 shown, in step 310, the device for identifying gestures receives the Euler angles of the finger phalanges measured by the sensors installed on the wearable device. Subsequently, proceed to step 320, and the device for identifying gestures generates the first vector J r ={J r (1), J r (2),…J r(i),…,J r (n)}, where i is the index of the finger joint, n is the number of finger joint feature parameters used to recognize gestures, and J r (i) represents the bending angle of the i-th finger joint or the angle between the phalanges on both sides of the joint.

[0106] Then, the process proceeds to step 330, where the apparatus for identifying a gesture determines a first vector J r Similarly, for each gesture to be recognized, it has a corresponding reference vector J for the first vector. k ={J k (1),J k (2),…J k (i),…,J k (n)}, where J k (i) represents the i-th component of the reference vector.

[0107] It should be noted that for each gesture to be recognized, there is a corresponding reference vector for the first vector, so the first vector has multiple similarities, each similarity corresponding to a gesture type.

[0108] Preferably, for the kth gesture type, the first vector J can be used r The corresponding reference vector J k The Euclidean distance between them is used to measure their similarity. Specifically, the similarity S can be calculated using the above formula (1): k1 .

[0109] Then, the process proceeds to step 340 to normalize the multiple similarities determined in step 330 to obtain a normalized similarity with a value range between 0 and 1.

[0110] Then, the process proceeds to step 350 to recognize the gesture according to the normalized similarity.

[0111] The normalized similarity S' of the first vector relative to the k-th gesture type k1 For example, the following rules can be used to determine whether the currently detected gesture is the kth gesture type:

[0112] If S'1<0.05, the currently detected gesture is determined to be highly similar to the k-th gesture type, if 0.05≤S'1<0.1, the currently detected gesture is determined to be moderately similar to the k-th gesture type, if 0.1≤S'1<0.15, the currently detected gesture is determined to be lowly similar to the k-th gesture type, and if S'1≥0.15, the currently detected gesture is determined to be dissimilar.

[0113] Figure 4The present invention is a device for recognizing gestures according to another embodiment of the present invention.

[0114] Figure 4 The device 40 shown includes a first module 410, a second module 420 and a third module 430. In this embodiment, the first module 410 is used to generate a first vector associated with the bending angle of one or more finger joints and a second vector associated with the spatial orientation of the palm, the second module 420 is used to determine the similarity between the first vector and the second vector and the respective reference vectors, and the third module 430 is used to recognize the gesture according to the similarity.

[0115] Figure 5 The present invention is a device for recognizing gestures according to another embodiment of the present invention.

[0116] Figure 5 The device 50 shown includes a first module 510, a second module 520 and a third module 530. In this embodiment, the first module 510 is used to generate a first vector associated with the bending angle of one or more finger joints, the second module 520 is used to determine the similarity between the first vector and a reference vector, and the third module 530 is used to recognize a gesture based on the similarity.

[0117] Figure 6 The present invention is a schematic block diagram of a device for recognizing gestures according to another embodiment of the present invention.

[0118] Figure 6 The device 60 for identifying gestures shown in the figure comprises a memory 610, a processor 620, and a computer program 630 stored in the memory 610 and executable on the processor 620, wherein executing the computer program 630 can implement the above-mentioned Figure 2 and 3 The method for recognizing gestures.

[0119] Figure 7 The present invention is a schematic block diagram of a wearable device according to another embodiment of the present invention.

[0120] like Figure 7 As shown, the wearable device 70 of this embodiment includes a first sensor 710, a second sensor 720 and a device 730 for identifying gestures. In this embodiment, the first sensor 710 is set on the wearable device in an area corresponding to the phalanges near the finger joints to obtain the bending angle of the finger joints, and the second sensor 720 is set on the wearable device in an area corresponding to the back of the hand or the palm to obtain the spatial orientation. The device 730 for identifying gestures can be combined with Figure 2 The device is implemented.

[0121] Figure 8The present invention is a schematic block diagram of a wearable device according to another embodiment of the present invention.

[0122] like Figure 8 As shown, the wearable device 80 of this embodiment includes a first sensor 810 and a device 820 for identifying gestures. In this embodiment, the first sensor 810 is disposed on the wearable device in an area corresponding to the phalanges near the finger joints to obtain the bending angles of the finger joints. The device 820 for identifying gestures can be combined with Figure 3 The device is implemented.

[0123] exist Figure 7 and 8 In the illustrated embodiment, wearable devices 70 and 80 may be wearable gloves.

[0124] Fig. 9 The present invention is a schematic block diagram of a system for recognizing gestures according to another embodiment of the present invention.

[0125] like Fig. 9 As shown, the system 90 for identifying gestures in this embodiment includes a wearable device 910 and a computing device 920. In this embodiment, the wearable device 910 includes a first sensor 911 and a second sensor 912, wherein the first sensor 911 is arranged on the wearable device in an area corresponding to the phalanges near the finger joints to obtain the bending angles of the finger joints, and the second sensor 912 is arranged on the wearable device in an area corresponding to the back of the hand or the palm to obtain the spatial orientation of the palm.

[0126] and Figure 7 and 8 The difference from the embodiment shown is that in this embodiment, the gesture recognition process is completed by a computing device 920 located outside the wearable device. Fig. 9 The computing device 920 includes a memory 921, a processor 922, and a computer program 923 stored in the memory 921 and executable on the processor 922, wherein the processor 922 is coupled to the first sensor 911 and the second sensor 912 to obtain data on the bending angle of the finger joints and the spatial orientation of the palm, and executes the computer program 923 to implement the above Figure 2 The method for recognizing gestures.

[0127] Fig.10 The present invention is a schematic block diagram of a system for recognizing gestures according to another embodiment of the present invention.

[0128] like Fig.10As shown, the system 100 for identifying gestures in this embodiment includes a wearable device 1010 and a computing device 1020. In this embodiment, the wearable device 1010 includes a first sensor 1011, which is arranged on the wearable device in an area corresponding to the phalanges near the finger joints to obtain the bending angle of the finger joints.

[0129] and Figure 7 and 8 The difference from the embodiment shown is that in this embodiment, the gesture recognition process is completed by the computing device 1020 located outside the wearable device. Fig.10 The computing device 1020 includes a memory 1021, a processor 1022, and a computer program 1023 stored in the memory 1021 and executable on the processor 1022, wherein the processor 1022 is coupled to the first sensor 1011 to obtain data of the bending angle of the finger joint, and executes the computer program 1023 to implement the above Figure 3 The method for recognizing gestures.

[0130] exist Fig. 9 and 10 In the illustrated embodiment, the wearable devices 910 and 1010 may be wearable gloves, and the computing devices 920 and 1020 may be personal computers, tablet computers, mobile phones, personal digital assistants, and the like.

[0131] According to another aspect of the present invention, there is also provided a computer readable storage medium on which a computer program is stored. When the program is executed by a processor, the above Figure 2 and 3 The method for recognizing gestures.

[0132] The embodiments and examples set forth herein are provided to best illustrate embodiments according to the present technology and its specific applications, and thereby enable those skilled in the art to make and use the present invention. However, those skilled in the art will appreciate that the above description and examples are provided for ease of illustration and example only. The description set forth is not intended to cover all aspects of the present invention or to limit the invention to the precise form disclosed.

[0133] In view of the foregoing, the scope of the present disclosure is determined by the following claims.

Claims

1. A method for identifying a gesture, characterized in that: It includes the following steps: generating a first vector associated with a bending angle of one or more finger joints and a second vector associated with a spatial orientation of a palm; determining similarities of the first vector and the second vector to respective ones of a plurality of reference vectors, wherein each reference vector corresponds to one of a plurality of gesture types; Normalizing each determined similarity to obtain a normalized similarity with a value range between 0 and 1; and Recognizing the gesture according to the normalized similarity includes: Determining the similarity between the first vector and the second vector and the respective reference vectors according to a value range to which each of the normalized similarities belongs, wherein the value range corresponds to one of a plurality of types of similarity, the types including high similarity, medium similarity, low similarity and dissimilarity; and For one of the multiple gesture types, determining whether the gesture to be recognized is one of the gesture types based on the similarity between the first vector and the second vector and a reference vector associated with the one of the gesture types, Wherein, the bending angle and spatial orientation are obtained by using sensors, Wherein, the second vector is in the form of a quaternion, Among them, for the similarity between the first vector and the corresponding reference vector, the value range less than 0.05 corresponds to high similarity, the value range greater than or equal to 0.05 and less than 0.1 corresponds to moderate similarity, the value range greater than or equal to 0.1 and less than 0.15 corresponds to low similarity, and the value range greater than or equal to 0.15 corresponds to dissimilarity, Among them, for the similarity between the second vector and the corresponding reference vector, the value range greater than 0.95 corresponds to high similarity, the value range less than or equal to 0.95 and greater than 0.90 corresponds to moderate similarity, the value range less than or equal to 0.90 and greater than 0.85 corresponds to low similarity, and the value range less than or equal to 0.85 corresponds to dissimilarity.

2. The method of claim 1, wherein: The bending angle is obtained by setting a sensor on an area of ​​the wearable device corresponding to the phalange near the finger joint.

3. The method of claim 1, wherein: The spatial orientation is obtained by arranging a sensor on the wearable device in an area corresponding to the back of the hand or the palm.

4. The method of claim 1, wherein: The similarity between the first vector and the corresponding reference vector is measured by the Euclidean distance between the two, and the similarity between the second vector and the corresponding reference vector is measured by the angle between the two.

5. A device for identifying gestures, characterized in that: Include: A first module for generating a first vector associated with a bending angle of one or more finger joints and a second vector associated with a spatial orientation of a palm; a second module for determining similarities between the first and second vectors and respective reference vectors, wherein each reference vector corresponds to one of the plurality of gesture types; Normalizing each determined similarity to obtain a normalized similarity with a value range between 0 and 1; and A third module is used to recognize a gesture according to the normalized similarity, including determining the similarity between the first vector and the second vector and their respective reference vectors according to a value range to which the normalized similarity belongs, wherein the value range corresponds to one of multiple types of similarity, wherein the types include high similarity, medium similarity, low similarity and dissimilarity, Wherein, the bending angle and spatial orientation are obtained by using sensors, Wherein, the second vector is in the form of a quaternion, Among them, for the similarity between the first vector and the corresponding reference vector, the value range less than 0.05 corresponds to high similarity, the value range greater than or equal to 0.05 and less than 0.1 corresponds to moderate similarity, the value range greater than or equal to 0.1 and less than 0.15 corresponds to low similarity, and the value range greater than or equal to 0.15 corresponds to dissimilarity, Among them, for the similarity between the second vector and the corresponding reference vector, the value range greater than 0.95 corresponds to high similarity, the value range less than or equal to 0.95 and greater than 0.90 corresponds to moderate similarity, the value range less than or equal to 0.90 and greater than 0.85 corresponds to low similarity, and the value range less than or equal to 0.85 corresponds to dissimilarity.

6. A device for identifying gestures, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The program is executed to implement the method according to any one of claims 1 to 4.

7. A wearable device comprising: A first sensor is disposed on the wearable device in an area corresponding to a phalanx near a finger joint to obtain a bending angle of the finger joint; A second sensor is disposed on the wearable device in an area corresponding to the back of the hand or the palm to obtain the spatial orientation of the palm; The device for recognizing gestures comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the following steps are implemented by executing the program: generating a first vector associated with a bending angle of one or more finger joints and a second vector associated with a spatial orientation of a palm; determining similarities of the first vector and the second vector to respective reference vectors, wherein each reference vector corresponds to one of a plurality of gesture types; Normalizing each determined similarity to obtain a normalized similarity with a value range between 0 and 1; and Recognizing a gesture according to the normalized similarity includes determining the similarity between the first vector and the second vector and respective reference vectors according to a value range to which the normalized similarity belongs, wherein the value range corresponds to one of a plurality of types of similarity, wherein the types include high similarity, medium similarity, low similarity and dissimilarity, Wherein, the second vector is in the form of a quaternion, Among them, for the similarity between the first vector and the corresponding reference vector, the value range less than 0.05 corresponds to high similarity, the value range greater than or equal to 0.05 and less than 0.1 corresponds to moderate similarity, the value range greater than or equal to 0.1 and less than 0.15 corresponds to low similarity, and the value range greater than or equal to 0.15 corresponds to dissimilarity, Among them, for the similarity between the second vector and the corresponding reference vector, the value range greater than 0.95 corresponds to high similarity, the value range less than or equal to 0.95 and greater than 0.90 corresponds to moderate similarity, the value range less than or equal to 0.90 and greater than 0.85 corresponds to low similarity, and the value range less than or equal to 0.85 corresponds to dissimilarity.

8. The wearable device according to claim 7, wherein: The wearable device is a wearable glove.

9. A system for recognizing gestures, comprising: Wearable devices, including: A first sensor is disposed on the wearable device in an area corresponding to a phalanx near a finger joint to obtain a bending angle of the finger joint; A second sensor is disposed on the wearable device in an area corresponding to the back of the hand or the palm to obtain the spatial orientation of the palm; A computing device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the following steps are implemented by executing the program: generating a first vector associated with a bending angle of one or more finger joints and a second vector associated with a spatial orientation of a palm; determining similarities of the first vector and the second vector to respective ones of a plurality of reference vectors, wherein each reference vector corresponds to one of a plurality of gesture types; Normalizing each determined similarity to obtain a normalized similarity with a value range between 0 and 1; and Recognizing a gesture according to the normalized similarity includes determining the similarity between the first vector and the second vector and respective reference vectors according to a value range to which the normalized similarity belongs, wherein the value range corresponds to one of a plurality of types of similarity, wherein the types include high similarity, medium similarity, low similarity and dissimilarity, Wherein, the second vector is in the form of a quaternion, Among them, for the similarity between the first vector and the corresponding benchmark vector, its value range less than 0.05 corresponds to high similarity, its value range greater than or equal to 0.05 and less than 0.1 corresponds to moderate similarity, its value range greater than or equal to 0.1 and less than 0.15 corresponds to low similarity, and its value range greater than or equal to 0.15 corresponds to dissimilarity, wherein, for the similarity between the second vector and the corresponding benchmark vector, its value range greater than 0.95 corresponds to high similarity, its value range less than or equal to 0.95 and greater than 0.90 corresponds to moderate similarity, its value range less than or equal to 0.90 and greater than 0.85 corresponds to low similarity, and its value range less than or equal to 0.85 corresponds to dissimilarity.

10. The system of claim 9, wherein: The wearable device is a wearable glove, and the computing device is one of a personal computer, a tablet computer, a mobile phone and a personal digital assistant.

11. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

12. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images