A gesture recognition human-computer interaction method and system for air purifiers

Abstract

The present application relates to the field of human-computer interaction technology, and in particular to a gesture recognition human-computer interaction method and system for air purifiers, the method comprising: taking the line connecting the center position of the key point of the head and the key point of the hand as the direction vector, calculating the willingness to interact at each moment, and comparing the willingness to interact at each moment with the preset value to obtain the interaction moment; calculating the similarity between the key point of the hand and the standard gesture at each interaction moment, and taking the sum of the product of the normalized willingness to interact and the similarity at each interaction moment as the first probability of the standard gesture; obtaining the change trend of the similarity at each interaction moment, and taking the sum of the change trend and the first probability as the second probability; realizing human-computer interaction according to the operation content of the standard gesture corresponding to the maximum value of the second probability. Through the technical solution of the present application, the accuracy of human-computer interaction can be improved, thereby improving the user experience.

Description

Technical Field

[0001] The present application relates to the field of human-computer interaction technology, and in particular to a gesture recognition human-computer interaction method and system for an air purifier. Background Art

[0002] Human-computer interaction refers to the technology that enables the machine to understand the user's interaction intention and perform corresponding feedback through communication between the user and the machine. It is applied in many fields and has broad development prospects. In the field of smart home, human-computer interaction between users and smart homes is often achieved through gesture recognition.

[0003] At present, the patent application document with application publication number CN117149046A discloses a method for adjusting the threshold value of interactive gesture intensity and an electronic device, the method comprising: counting the first number of times a first interactive gesture is recognized within a first time, the first interactive gesture conforms to the first interactive mode, and the interactive intensity value of the first interactive gesture is less than the first intensity value and greater than the second intensity value, and the second intensity value is less than the first intensity value; judging whether the first number is greater than the preset number threshold; if so, adjusting the interactive gesture intensity threshold of the first interactive mode. Wherein, the interactive matching degree detection comprises: judging whether the matching degree between the first sensor time series signal corresponding to the interactive gesture and the second sensor time series signal corresponding to the first interactive mode is greater than the first matching degree; the interactive intensity detection comprises: according to the determined first interactive mode, obtaining the interactive intensity value of the interactive gesture, judging whether the interactive intensity value of the interactive gesture is less than the first intensity value and greater than the second intensity value; wherein, when the first interactive mode is a tapping mode, the interactive intensity value is the tapping intensity; when the first interactive mode is a pressing mode, the interactive intensity value is the finger pressing intensity.

[0004] The above method can actively guide the user to adjust or actively adjust the interaction gesture strength threshold when the success rate of interaction gesture recognition is low, thereby improving the success rate of interaction gesture recognition. However, in the process of human-computer interaction, when it is detected that the user has an interaction gesture, it does not necessarily mean that the user has the intention to interact. Ignoring the detection of the user's intention to interact will lead to miscontrol, reduce the accuracy of human-computer interaction, and the user experience is not high. Summary of the invention

[0005] In order to solve the technical problems of inaccurate human-computer interaction and poor user experience, the present application provides a gesture recognition human-computer interaction method and system for an air purifier, which can improve the accuracy of human-computer interaction and thus improve the user experience.

[0006] In a first aspect of the present application, a gesture recognition human-computer interaction method for an air purifier is provided, the interaction method comprising: taking a line connecting a head key point and a center position of a hand key point as a direction vector, calculating the interaction willingness at each moment, and comparing the interaction willingness at each moment with a preset value to obtain an interaction moment, wherein the interaction willingness is negatively correlated with the angle between the direction vector and the preset direction, and positively correlated with the decreasing speed of the angle, and the preset direction is a direction vector pointing from the head key point to a machine position point; calculating the similarity between the hand key point at each interaction moment and any standard gesture, and taking the sum of the products of the normalized interaction willingness and the similarity at each interaction moment as a first probability of the standard gesture; obtaining a changing trend of the similarity at each interaction moment, and taking the sum of the changing trend and the first probability as a second probability; and realizing human-computer interaction according to the operation content of the standard gesture corresponding to the maximum value of the second probability.

[0007] Preferably, before taking the line connecting the center positions of the head key points and the hand key points as the direction vector, the interaction method further includes: subtracting the real-time image at any moment from the background image to obtain a difference map, wherein the background image is an image captured when there is no motion area; in response to the average pixel value in the difference map being greater than the pixel threshold, obtaining the coordinate positions of the head key points and the hand key points at multiple moments after the moment.

[0008] Preferably, the center position of the hand key point is the average coordinate of the coordinate positions of each hand key point.

[0009] Preferably, the time Willingness to interact satisfy:

[0010] ;in, and Separately for the moment and time The angle between the direction vector and the preset direction, for Activation function.

[0011] Preferably, the interaction willingness at each moment is compared with a preset value to obtain the interaction moment, including: in response to the interaction willingness at any moment being greater than the preset value, the moment is taken as the starting moment; the first moment after the starting moment when the interaction willingness is greater than the preset value is taken as the ending moment; counting the number of moments between the starting moment and the ending moment, and in response to the number of moments being greater than a number threshold, each moment between the starting moment and the ending moment is taken as an interaction moment, otherwise, no interaction moment is detected.

[0012] Preferably, the interaction willingness at each moment is compared with a preset value to obtain the interaction moment, including: in response to the interaction willingness at any moment being greater than the preset value, the moment is taken as the interaction moment; the interaction willingness at each moment is compared with the preset value in turn, and the interaction moment is continuously obtained until the interaction moment is not obtained after a preset time interval.

[0013] Preferably, the interaction time Hand key points and standard gestures Similarity The methods for obtaining include: Among the hand key points, the line connecting the two selected hand key points is used as the reference line; the rotation moment After the hand key points are aligned so that the angle between the reference line and the horizontal direction is equal to the preset angle, the similarity is calculated. , the similarity satisfy:

[0014] , is the number of key points of the hand, and Respectively indicate time and standard gestures Middle hand key points The coordinate position of and Respectively indicate time and standard gestures The center position of the middle hand key point.

[0015] Preferably, standard gestures The first probability satisfy:

[0016] , is the number of interaction moments, The willingness to interact at the interaction moment, is the sum of the willingness to interact at all interaction moments, For interactive moments Hand key points and standard gestures The similarity.

[0017] Preferably, standard gestures The method for obtaining the changing trend of is:

[0018] ; and is a set of similarity pairs, which are interaction moments and interactive moments Hand key points and standard gestures similarity; Indicates that the interaction moment is satisfied All similarity pairs of is the mean value function.

[0019] In a second aspect of the present application, a gesture recognition human-computer interaction system for an air purifier is provided, comprising a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, a gesture recognition human-computer interaction method for an air purifier described in the first aspect of the present application is implemented.

[0020] The technical solution of this application has the following beneficial technical effects:

[0021] The image is collected by an image acquisition device deployed on the air purifier to obtain the head key points and hand key points in the image, and the line connecting the center positions of the head key points and the hand key points is used as the direction vector. The interaction willingness of the user and the air purifier at each moment is judged according to the angle between the direction vector and the preset direction and the decreasing speed of the angle, and then the interaction moment is determined by comparing the interaction willingness with the preset value. The hand key points at the interaction moment can be used for human-computer interaction to avoid the influence of the hand key points at the non-interaction moment on the human-computer interaction. Further, for any standard gesture, the similarity between the hand key points at each interaction moment and the standard gesture is calculated, and the sum of the products of the normalized interaction willingness and the similarity at each interaction moment is used as the first probability of the standard gesture, and the similarity of the standard gesture is analyzed in time series, and the first probability is adjusted according to the changing trend of the similarity at each interaction moment to obtain the second probability of the standard gesture. The second probability of each standard gesture is obtained by comprehensively combining the similarity, the interaction willingness at the interaction moment, and the changing trend of the similarity, so as to improve the accuracy of gesture recognition, and realize human-computer interaction according to the operation content of the standard gesture corresponding to the maximum value of the second probability, thereby improving the accuracy of human-computer interaction and thus improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present application will become easy to understand. In the accompanying drawings, several embodiments of the present application are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0023] Figure 1 is a structural diagram of an air purifier according to an embodiment of the present application;

[0024] Figure 2 is a flow chart of a gesture recognition human-computer interaction method for an air purifier according to an embodiment of the present application;

[0025] Figure 3is a schematic diagram of key points of a hand according to an embodiment of the present application;

[0026] Figure 4 is a schematic diagram of a reference line in an embodiment of the present application;

[0027] Figure 5 It is a structural block diagram of a gesture recognition human-computer interaction system for an air purifier according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0029] It should be understood that when the terms "first", "second", etc. are used in the claims, specification and drawings of the present application, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the specification and claims of the present application indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their collections.

[0030] According to a first aspect of the present application, the present application provides a gesture recognition human-computer interaction method for an air purifier. Figure 1 It is a structural diagram of an air purifier according to an embodiment of the present application; an image acquisition device 201 is deployed on the air purifier for acquiring image information in front of the air purifier, and the image acquisition device is a pinhole camera or a high-definition camera; the air purifier is also provided with a display screen 202 for displaying the operating status of the air purifier, and the operating status includes motor gear and / or environmental information, and the environmental information includes PM2.5, temperature and humidity.

[0031] Figure 2 is a flow chart of a gesture recognition human-computer interaction method for an air purifier according to an embodiment of the present application. Figure 2 As shown, the gesture recognition human-computer interaction method for an air purifier includes steps S101 to S104, which are described in detail below.

[0032] S101, taking the line connecting the center positions of the head key points and the hand key points as the direction vector, calculating the interaction willingness at each moment, and comparing the interaction willingness at each moment with the preset value to obtain the interaction moment, the interaction willingness is negatively correlated with the angle between the direction vector and the preset direction, and positively correlated with the decreasing speed of the angle, and the preset direction is the direction vector pointing from the head key point to the machine position point.

[0033] In one embodiment, the image acquisition device deployed on the air purifier can acquire real-time images at each moment; the real-time image at any moment is subtracted from the background image to obtain a difference map, wherein the background image is an image acquired when there is no motion area; in response to the average pixel value in the difference map being greater than a pixel threshold, indicating that there is a motion area in the real-time image at that moment, the coordinate positions of the key points of the head and the key points of the hands at multiple moments after the moment are acquired. The pixel threshold is 1.

[0034] Among them, existing human key point detection networks such as OpenPose, HRNet or MediaPipe can be used for implementation. Figure 3 , is a schematic diagram of the hand key points of an embodiment of the present application, 21 hand key points can be obtained at each moment, and each hand key point corresponds to a coordinate position, the center position of the hand key point is the average coordinate of the coordinate position of each hand key point, and the center position of the hand key point can be regarded as the center of the palm, and the line connecting the head key point and the center position of the hand key point is taken as the direction vector, and the modulus length of the direction vector is 1, which is used to reflect the extending direction of the palm.

[0035] In another embodiment, the coordinate positions of the key points of the head and the key points of the hand in the real-time image at each moment may also be obtained, which is not limited in this application.

[0036] In one embodiment, when the user reaches out to the air purifier, it indicates that the user has a willingness to interact with the air purifier. At the same time, when the user's extended direction quickly points to the air purifier, it also indicates that the user has a willingness to interact with the air purifier. Since the position of the image acquisition device is fixed, the machine position point can be pre-marked in the image. The machine position is the position of the air purifier in the image, and the direction vector pointing from the head key point to the machine position point is used as the preset direction. The smaller the angle between the direction vector and the preset direction, the closer the extended direction of the palm is to the air purifier, and the stronger the user's willingness to interact with the air purifier; the greater the decreasing speed of the angle between the direction vector and the preset direction, the faster the extended direction of the palm approaches the air purifier, and the stronger the user's willingness to interact with the air purifier. Therefore, the willingness to interact is negatively correlated with the angle between the direction vector and the preset direction, and positively correlated with the decreasing speed of the angle.

[0037] Specifically, the time Willingness to interact satisfy:

[0038] ;in, and Separately for the moment and time The angle between the direction vector and the preset direction, for Activation function. Among them, To reflect the moment The falling speed of the angle between the direction vector and the preset direction, The larger it is, the faster the decline rate is, and the stronger the user's willingness to interact with the air purifier is.

[0039] In other embodiments, the time Willingness to interact You can also: , and Separately for the moment and time The angle between the direction vector and the preset direction; satisfy the interaction intention With the moment The angle between the direction vector and the preset direction Negatively correlated with the rate of decline A positive correlation is sufficient.

[0040] In one embodiment, after obtaining the interaction willingness at each moment, the interaction moment when the user and the air purifier have behavior can be determined based on the interaction willingness. Specifically, the interaction willingness at each moment is compared with the preset value to obtain the interaction moment, including: in response to the interaction willingness at any moment being greater than the preset value, the moment is taken as the starting moment; the first moment after the starting moment when the interaction willingness is greater than the preset value is taken as the ending moment; the number of moments between the starting moment and the ending moment is counted, and the number of response moments is greater than the number threshold, and each moment between the starting moment and the ending moment is taken as the interaction moment, otherwise, no interaction moment is detected. Among them, the preset value is 0.8.

[0041] It should be noted that the quantity threshold is 3. When the number of moments between the start moment and the end moment is less than the quantity threshold, it means that the duration of the user's willingness to interact with the air purifier is short and does not meet the minimum duration requirement. Multiple moments between the start moment and the end moment are not regarded as interaction moments. This can avoid miscontrol of the air purifier and improve the accuracy of human-computer interaction.

[0042] In another embodiment, comparing the interaction willingness at each moment with a preset value to obtain the interaction moment includes: in response to the interaction willingness at any moment being greater than the preset value, taking the moment as the interaction moment; sequentially comparing the interaction willingness at each moment with the preset value, and continuously obtaining the interaction moment until the interaction moment is not obtained after a preset time interval, where the preset time interval is 1 second.

[0043] In this way, the interaction moment when the user has the intention to interact with the air purifier is obtained, and the human-computer interaction between the user and the air purifier can be realized based on the gestures at the interaction moment.

[0044] S102, calculating the similarity between the key points of the hand at each interaction moment and any standard gesture, and taking the sum of the products of the normalized interaction intention and the similarity at each interaction moment as the first probability of the standard gesture.

[0045] In one embodiment, the interaction moment is the moment when there is a willingness to interact between the user and the air purifier, and the willingness to interact is different at different moments. The air purifier will pre-set multiple standard gestures, and different standard gestures correspond to different operation contents. For example, the operation content represented by gesture 1 is "adjust the motor gear by one gear", and the operation content represented by gesture 2 is "adjust the motor gear by one gear".

[0046] In one embodiment, the interaction time Hand key points and standard gestures Similarity The methods for obtaining include: Among the hand key points, the line connecting the two selected hand key points is used as the reference line; the rotation moment After the hand key points are aligned so that the angle between the reference line and the horizontal direction is equal to the preset angle, the similarity is calculated. , the similarity satisfy:

[0047] , is the number of key points of the hand, and Respectively indicate time and standard gestures Middle hand key points The coordinate position of and Respectively indicate time and standard gestures The center position of the middle hand key point.

[0048] Among them, the time and standard gestures The number of key points of the hand is the same as that of Figure 4, is a schematic diagram of a reference line of an embodiment of the present application, taking hand key point 12 and hand key point 0 as selected hand key points, connecting hand key point 12 and hand key point 0 to obtain a reference line, the angle between the reference line and the horizontal direction can reflect the direction of the hand; in calculating the similarity Before, the rotation moment The hand key points are set so that the angle between the reference line and the horizontal direction is equal to the preset angle, which can avoid the influence of the hand direction on the similarity and improve the similarity. accuracy.

[0049] In one embodiment, the hand key points and standard gestures are obtained at each interaction moment. After the similarity of The first probability , standard gesture The first probability satisfy: , is the number of interaction moments, The willingness to interact at the interaction moment, is the sum of the willingness to interact at all interaction moments, For interactive moments Hand key points and standard gestures The similarity.

[0050] in, For interactive moments Normalized willingness to interact.

[0051] In this way, a first probability of each standard gesture is obtained, and the larger the first probability is, the greater the possibility that the user makes the corresponding standard gesture is.

[0052] S103: Obtain a change trend of the similarity of each interaction moment, and take the sum of the change trend and the first probability as a second probability.

[0053] In one embodiment, during the interaction between the user and the air purifier, a corresponding gesture will be made from the first interaction moment, and the corresponding gesture will gradually approach the standard gesture, so that the air purifier can perform the operation content corresponding to the standard gesture. Therefore, the similarity of the standard gesture is analyzed in time sequence to obtain the similarity change trend. If the similarity of a standard gesture changes in time sequence and the more obvious the growth trend is, the greater the possibility that the user makes the standard gesture. The first probability of the standard gesture can be adjusted using the similarity change trend to obtain a more accurate gesture recognition result.

[0054] Specifically, the second probability of the standard gesture is equal to the sum of the change trend and the first probability. The second probability satisfy: , Standard gesture The changing trend of Standard gesture The first probability.

[0055] Among them, standard gestures The method for obtaining the changing trend of is:

[0056] ; and is a set of similarity pairs, which are interaction moments and interactive moments Hand key points and standard gestures similarity; Indicates that the interaction moment is satisfied All similarity pairs of is the mean value function.

[0057] When the standard gesture When the similarity of the time series changes in an increasing trend, is greater than 0, and the more obvious the growth trend is, The larger the value, the larger the standard gesture. When the similarity of is decreasing in time series, Less than 0, and the downward trend is more obvious, The smaller the value.

[0058] S104, realizing human-computer interaction according to the operation content of the standard gesture corresponding to the maximum value of the second probability.

[0059] In one embodiment, the standard gesture corresponding to the maximum value of the second probability is taken as the gesture recognition result, and the operation content corresponding to the gesture recognition result is executed to complete the human-computer interaction with the air purifier.

[0060] The technical principles and implementation details of the gesture recognition human-computer interaction method for an air purifier of the present application are introduced above through specific embodiments. The image is collected by an image acquisition device deployed on the air purifier to obtain the head key points and hand key points in the image, and the line connecting the center positions of the head key points and the hand key points is used as the direction vector. The interaction willingness of the user and the air purifier at each moment is judged according to the angle between the direction vector and the preset direction and the decreasing speed of the angle, and then the interaction moment is determined by comparing the interaction willingness with the preset value. The hand key points at the interaction moment can be used for human-computer interaction to avoid the influence of the hand key points at the non-interaction moment on the human-computer interaction. Further, for any standard gesture, the similarity between the hand key points at each interaction moment and the standard gesture is calculated, and the sum of the products of the normalized interaction willingness and the similarity at each interaction moment is used as the first probability of the standard gesture, and the similarity of the standard gesture is analyzed in time series, and the first probability is adjusted according to the changing trend of the similarity at each interaction moment to obtain the second probability of the standard gesture. The second probability of each standard gesture is obtained by comprehensively combining the similarity, the interaction willingness at the interaction moment, and the changing trend of the similarity, so as to improve the accuracy of gesture recognition, and realize human-computer interaction according to the operation content of the standard gesture corresponding to the maximum value of the second probability, thereby improving the accuracy of human-computer interaction and thus improving the user experience.

[0061] According to the second aspect of the present application, the present application also provides a gesture recognition human-computer interaction system for an air purifier. Figure 5 is a structural block diagram of a gesture recognition human-computer interaction system for an air purifier according to an embodiment of the present application. Figure 5 As shown, the system 50 includes a processor and a memory, the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a gesture recognition human-computer interaction method for an air purifier according to the first aspect of the present application is implemented. The system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, whose settings and functions are known in the art, and therefore will not be described in detail here.

[0062] In the present application, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of a device or accessible or connectable to a device. Any application or module described in this application may be implemented using computer-readable / executable instructions that may be stored or otherwise maintained by such a computer-readable medium.

[0063] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0064] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the patent application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent application shall be subject to the attached claims.

Claims

1. A gesture recognition human-computer interaction method for an air purifier, characterized in that: The interaction method comprises: The line connecting the center of the key points of the head and the key points of the hands is used as the direction vector to calculate the interaction intention at each moment, and compare the interaction intention at each moment with the preset value to obtain the interaction moment. Willingness to interact satisfy: ;in, and Separately for the moment and time The angle between the direction vector and the preset direction, for Activation function, the preset direction is a direction vector pointing from the head key point to the machine position point; Calculate the similarity between the key points of the hand at each interaction moment and any standard gesture, and take the sum of the products of the normalized interaction intention and the similarity at each interaction moment as the first probability of the standard gesture; Obtaining a change trend of the similarity at each interaction moment, and taking the sum of the change trend and the first probability as a second probability; Human-computer interaction is achieved according to the operation content of the standard gesture corresponding to the maximum value of the second probability.

2. A gesture recognition human-computer interaction method for an air purifier according to claim 1, characterized in that: Before taking the line connecting the center positions of the head key point and the hand key point as the direction vector, the interaction method further includes: Subtract the real-time image at any time from the background image to obtain a difference map, wherein the background image is an image collected when there is no motion area; In response to the average pixel value in the difference map being greater than the pixel threshold, the coordinate positions of the head key points and the hand key points at multiple moments after the moment are obtained.

3. The gesture recognition human-computer interaction method for an air purifier according to claim 1, characterized in that: The center position of the hand key points is the average coordinate of the coordinate positions of each hand key point.

4. The gesture recognition human-computer interaction method for an air purifier according to claim 1, characterized in that: Compare the interaction intention at each moment with the preset value to obtain the interaction moment, including: In response to the interaction willingness at any time being greater than a preset value, taking the time as the starting time; The first moment after the start time when the interaction willingness is greater than the preset value is taken as the end time; The number of moments between the start time and the end time is counted. If the number of response moments is greater than the number threshold, each moment between the start time and the end time is regarded as an interaction moment. Otherwise, no interaction moment is detected.

5. The gesture recognition human-computer interaction method for an air purifier according to claim 1, characterized in that: Compare the interaction intention at each moment with the preset value to obtain the interaction moment, including: In response to the interaction willingness at any time being greater than a preset value, the time is regarded as the interaction time; The interaction intention at each moment is compared with the preset value in turn, and the interaction moment is continuously obtained until no interaction moment is obtained after a preset time interval.

6. The method for human-computer interaction of gesture recognition for an air purifier according to claim 1, characterized in that: Interactive Moments Hand key points and standard gestures Similarity The methods for obtaining include: At the moment Among the hand key points, the line connecting the two selected hand key points is used as the reference line; the rotation moment After the hand key points are aligned so that the angle between the reference line and the horizontal direction is equal to the preset angle, the similarity is calculated. , the similarity satisfy: , is the number of key points of the hand, and Respectively indicate time and standard gestures Middle hand key points The coordinate position of and Respectively indicate time and standard gestures The center position of the middle hand key point.

7. A gesture recognition human-computer interaction method for an air purifier according to claim 6, characterized in that: Standard gestures The first probability satisfy: , is the number of interaction moments, The willingness to interact at the interaction moment, is the sum of the willingness to interact at all interaction moments, For interactive moments Hand key points and standard gestures The similarity.

8. The gesture recognition human-computer interaction method for an air purifier according to claim 1, characterized in that: Standard gestures The method for obtaining the changing trend of is: ; and is a set of similarity pairs, which are interaction moments and interactive moments Hand key points and standard gestures similarity; Indicates that the interaction moment is satisfied All similarity pairs of is the mean value function.

9. A gesture recognition human-computer interaction system for an air purifier, characterized in that: It comprises a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a gesture recognition human-computer interaction method for an air purifier according to any one of claims 1 to 8 is implemented.

Images