CN116597473B - Gesture recognition method, device, equipment and storage medium - Google Patents
Gesture recognition method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN116597473B CN116597473B CN202310553653.1A CN202310553653A CN116597473B CN 116597473 B CN116597473 B CN 116597473B CN 202310553653 A CN202310553653 A CN 202310553653A CN 116597473 B CN116597473 B CN 116597473B
- Authority
- CN
- China
- Prior art keywords
- standard
- data
- gesture
- hand data
- hand
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000012545 processing Methods 0.000 claims abstract description 14
- 230000001174 ascending effect Effects 0.000 claims description 17
- 238000012790 confirmation Methods 0.000 claims description 14
- 238000005452 bending Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 4
- 230000003993 interaction Effects 0.000 abstract description 2
- 230000008859 change Effects 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 210000001145 finger joint Anatomy 0.000 description 2
- 210000003205 muscle Anatomy 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 210000000988 bone and bone Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/107—Static hand or arm
- G06V40/11—Hand-related biometrics; Hand pose recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
- G06F16/51—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Human Computer Interaction (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The application relates to a gesture recognition method, a device, equipment and a storage medium, which are applied to the field of gesture interaction, wherein the method comprises the following steps: controlling a hand data acquisition device to acquire hand data of a user; processing the hand data to obtain standard hand data; invoking a standard gesture model from a preset gesture model database; determining sample data of the standard gesture model, comparing the standard hand data with the sample data, and calculating a data difference value between the standard hand data and the sample data; judging whether the data difference value is within a preset reasonable difference value range or not; and if the standard hand data is positioned, judging that the standard hand data is successfully matched with the sample data. The technical effect that this application had is: instability in the gesture recognition process is reduced as much as possible.
Description
Technical Field
The present disclosure relates to the field of gesture interaction technologies, and in particular, to a gesture recognition method, device, apparatus, and storage medium.
Background
In computer science, gesture recognition is an issue of recognizing human gestures through mathematical algorithms. Gesture recognition may come from the motion of parts of a person's body, but generally refers to the motion of the face and hands.
Gesture recognition can be seen as a way of computing mechanisms to solve human language, building a richer bridge between the robot and the person than the original text user interface or even GUI (graphical user interface). The user can control or interact with the device using simple gestures without touching them. The core technology is gesture segmentation, gesture analysis and gesture recognition.
In carrying out the present application, the inventors have found that at least the following problems exist in this technology: generally, most gesture recognition methods used in the market use hand image data to extract hand contour points for recognition, and are greatly affected by factors such as the sharpness of hand images.
Disclosure of Invention
In order to reduce instability in a gesture recognition process as much as possible, the gesture recognition method, device, equipment and storage medium are provided.
In a first aspect, the present application provides a gesture recognition method, which adopts the following technical scheme: the method comprises the following steps: controlling a hand data acquisition device to acquire hand data of a user;
processing the hand data to obtain standard hand data;
invoking a standard gesture model from a preset gesture model database;
sample data of the standard gesture model is determined,
comparing the standard hand data with the sample data, and calculating a data difference value between the standard hand data and the sample data;
judging whether the data difference value is within a preset reasonable difference value range or not;
and if the standard hand data is positioned, judging that the standard hand data is successfully matched with the sample data.
Through the technical scheme, the hand data acquisition device is utilized by the gesture recognition system to acquire the current hand data of the user, the acquired hand data are compared with each sample data in the preset gesture model data, and the sample data matched with the current hand data are found, so that the effect of recognizing the gesture placed by the user is realized, compared with image recognition, the effect of recognizing the gesture placed by the user is not influenced by factors such as the definition of the hand image, and the instability in the gesture recognition process is reduced.
In a specific embodiment, the hand data includes at least: the key point coordinate values and the key point connection relations.
Through the technical scheme, the standard hand data for gesture recognition comprise all information of the hands of the user as much as possible, but the situation of data redundancy cannot occur, so that the time spent for gesture recognition is reduced while the accuracy of the gesture recognition result is ensured, and the efficiency of gesture recognition is further improved.
In a specific embodiment, the processing the hand data to obtain standard hand data specifically includes:
determining a key point connection relation in the hand data;
acquiring adjacent key point coordinate values according to the key point connection relation,
calculating a hand bending angle according to the coordinate values of the key points;
and marking the hand bending angle as standard hand data.
Through the technical scheme, the gesture recognition system processes the collected hand data of the user, so that the negative influence on the final gesture recognition result caused by the difference between the hand sizes of different users and the length of fingers is reduced, and the accuracy of the gesture recognition result is improved.
In a specific embodiment, after said if located, determining that said standard hand data matches said sample data successfully, further comprising:
recording the successful times of matching each standard gesture model in the gesture model database;
the step of retrieving the standard gesture model from a preset gesture model database specifically comprises the following steps:
according to the successful times of the matching corresponding to the standard gesture models, the standard gesture models in the gesture model database are arranged in ascending order, and each standard gesture model corresponds to a unique serial number;
and sequentially acquiring the standard gesture models according to the serial numbers.
According to the technical scheme, the gesture recognition system judges the standard gesture models commonly used by the user according to the successful matching times corresponding to each standard gesture model in the gesture model database, and the standard gesture models in the gesture model database are arranged in ascending order according to the successful matching times, so that the smaller the serial number value corresponding to the standard gesture model with more successful matching times is, the higher the serial number value is, and when the hand data is matched, the standard gesture model with more successful matching times is matched, so that the efficiency of gesture recognition is improved.
In a specific implementation manner, the standard gesture models in the gesture model database are arranged in ascending order according to the matching success times corresponding to the standard gesture models, and the method further includes:
in the process of ascending order arrangement of the standard gesture models, if the matching success times of a plurality of standard gesture models are equal, inquiring historical time information of the last successful matching of the standard gesture data;
and sequencing the standard gesture data according to the sequence of the historical time information.
According to the technical scheme, in the process of ascending arrangement of the standard gesture models in the gesture model database, if a plurality of standard gesture models have the same successful matching times, the gesture recognition system can check the recent use condition of the standard gesture models, the standard gesture models are ordered according to the index of the time value of the last successful matching, and the gesture recognition system also includes the stepwise change in the ordering index due to the stepwise change of the use condition of different gestures of a user, so that the final ordering result is more in accordance with the actual condition of the user, and the efficiency of gesture recognition is further improved.
In a specific embodiment, after said determining whether the data difference is within a preset reasonable difference range, the method further includes:
if not, a name confirmation instruction is sent to the user terminal;
receiving a name feedback instruction sent by a user terminal, wherein the name feedback instruction comprises a gesture model name;
establishing a correlation between the gesture model name and the standard hand data to generate a standard gesture model;
and storing the standard gesture model into a gesture model database.
Through the technical scheme, when the standard gesture model matched with the current hand data of the user does not exist in the gesture model database, the gesture recognition system can require the user to name the new gesture and store the new gesture into the gesture model database, so that standard gesture model resources in the gesture model database are enriched, and the situation that the subsequent hand data matching fails is reduced.
In a specific embodiment, if the user terminal is not located, the sending a name confirmation instruction to the user terminal specifically includes:
if not, the standard hand data is matched with the sample data again;
and when the continuous failure times of the standard hand data and the sample data reach a preset matching times threshold, sending a name confirmation instruction to the user terminal.
Through the technical scheme, when the hand data of the user is inconsistent with the standard gesture model in the gesture model database, the gesture recognition system can execute the matching process again, and when the number of times of repeated execution reaches a certain threshold value, the gesture recognition system can judge that the standard gesture model matched with the current hand data of the user does not exist in the gesture model database, repeated matching is conducted for multiple times, so that the situation of mismatching of the gesture recognition system is reduced, and the accuracy of gesture recognition is improved.
In a second aspect, the present application provides a gesture recognition apparatus, which adopts the following technical scheme: the device comprises:
the hand data acquisition module is used for controlling the hand data acquisition device to acquire hand data of a user;
the hand data processing module is used for processing the data format of the hand data to obtain standard hand data;
the gesture model retrieving module is used for retrieving a standard gesture model from a preset gesture model database;
a sample data determining module for determining sample data of the standard gesture model,
the data difference calculation module is used for comparing the standard hand data with the sample data and calculating a data difference between the standard hand data and the sample data;
the data difference value comparison module is used for judging whether the data difference value is in a preset reasonable difference value range or not;
and the gesture model matching module is used for judging that the standard hand data and the sample data are successfully matched if the gesture model is positioned.
In a third aspect, the present application provides a computer device, which adopts the following technical scheme: comprising a memory and a processor, said memory having stored thereon a computer program capable of being loaded by the processor and performing any of the gesture recognition methods described above.
In a fourth aspect, the present application provides a computer readable storage medium, which adopts the following technical solutions: a computer program is stored that can be loaded by a processor and that performs any of the gesture recognition methods described above.
In summary, the present application includes at least one of the following beneficial technical effects:
1. the gesture recognition system judges standard gesture models commonly used by a user according to successful matching times corresponding to each standard gesture model in the gesture model database, and the standard gesture models in the gesture model database are arranged in ascending order according to the successful matching times, so that the serial number value corresponding to the standard gesture model with more successful matching times is smaller, and when hand data matching is carried out, the standard gesture model with more successful matching times is preferentially matched, thereby being beneficial to improving the efficiency of gesture recognition;
2. in the process of ascending order arrangement of the standard gesture models in the gesture model database, if a plurality of standard gesture models have the same successful matching times, the gesture recognition system can check the recent use condition of the standard gesture models, and order the standard gesture models according to the index of the time value of the last successful matching, and the gesture recognition system also includes the stepwise change in the order index due to the stepwise change of the use condition of different gestures of the user, so that the final order result is more in accordance with the actual condition of the user, and the efficiency of gesture recognition is further improved.
Drawings
FIG. 1 is a flow chart of a gesture recognition method in an embodiment of the present application.
Fig. 2 is a schematic diagram of a hand recognition process in an embodiment of the present application.
FIG. 3 is a block diagram of a gesture recognition apparatus in an embodiment of the present application.
Reference numerals: 301. the hand data acquisition module; 302. a hand data processing module; 303. a gesture model calling module; 304. a sample data determination module; 305. a data difference calculation module; 306. a data difference comparison module; 307. and a gesture model matching module.
Detailed Description
The present application is described in further detail below in conjunction with figures 1-3.
The embodiment of the application discloses a gesture recognition method. The method is applied to a gesture recognition system, and program codes corresponding to the method are prestored in a control center of the gesture recognition system. In the application, the hand data acquisition device is a data glove, and a plurality of types of sensors are arranged in the data glove and used for detecting bending and position information of each key point of the hand of a user. In the present application, key points of the hand, including the finger joint gesture recognition system, may be developed based on mainstream game engines (e.g., unity3D, etc.) to provide a user friendly interface and a good interactive experience. Before the gesture recognition system is put into use, a user can input gesture data common to the user into the gesture recognition system in advance, specifically, the user needs to wear a hand data acquisition device, then bones and muscles of the hand are controlled to put out corresponding gestures, and in the application, the hand gestures mainly refer to static gestures.
The hand data acquisition device can capture the swung-out shape of the hand of the user and convert the swung-out shape into a data form. When a user inputs a gesture, the gesture recognition system records the gesture data and a gesture name corresponding to the gesture data, and in the application, definition of the gesture name can be in two forms, which are user-defined or generated by the gesture recognition system according to a preset naming rule. After the completion of the collection of the gesture data, the user associates the gesture with a certain operation instruction, so that when the user reappears the gesture, the gesture recognition system can recognize and execute the corresponding operation associated with the gesture.
As shown in fig. 1 and 2, the method comprises the steps of:
s10, controlling the hand data acquisition device to acquire hand data of a user.
Specifically, the user wears the hand data acquisition device, puts out the corresponding hand gesture by using the hand skeleton and the muscle, and starts the hand data acquisition device, and the hand data acquisition device acquires the hand data of the current user according to various preset sensors. Then, the hand data acquisition device will send the hand data that gathers to the control center of gesture recognition system in, in this embodiment, the data transmission mode between hand data acquisition device and the control center of gesture recognition system can be bluetooth transmission, infrared interface transmission, wireless transmission etc..
S20, processing the data format of the hand data to obtain standard hand data.
Specifically, because the model and the type of the hand data acquisition device are not unique, the data format of the hand data can be inconsistent, so that the gesture recognition system can convert the data format corresponding to the acquired hand data, so that all the hand data share one data format, for example, some hand data acquisition devices can define a new data type for storing the acquired hand data, and therefore, the data types defined by different hand data acquisition devices can be different, and after the hand data sent by the hand data acquisition device are received, the control center of the gesture recognition system can convert the data format corresponding to the acquired hand data, unify the data format of the hand data processed later, and in this way, the hand data processing for searching by the gesture model can be more convenient and flexible, and meanwhile, the consistency and reliability of the data can be ensured.
After the adjustment of the hand data is completed, the gesture recognition system will first determine the connection relationship of the key points in the hand data, obtain the coordinate values of the adjacent key points according to the connection relationship of the key points, generally, obtain three interconnected key point coordinate values at a time, for example, the finger head, the finger tail and the first joint point of the finger near the finger, for convenience of description, the first joint point of the finger near the finger is taken as the midpoint, and the three key points are the interconnected key points. And then calculating the bending angle of the finger according to the coordinate value of the key point, based on the continuous example, taking the middle point of the middle finger as a base point, constructing vectors of the finger head and the middle point and vectors of the finger tail and the middle point, calculating the finger section of the finger head and the middle point and the included angle between the finger tail and the finger section of the middle point by using the two vectors, and marking the bending angles of different fingers and the palm as standard hand data as the bending angle of the middle finger.
S30, retrieving a standard gesture model from a preset gesture model database.
Specifically, gesture data input into the gesture recognition system by the user in advance is stored into a preset gesture model database by the gesture recognition system, any group of gesture data in the gesture model database is a standard gesture model, and in this embodiment, data content included in the standard gesture model includes standard hand data, gesture names and operation instructions corresponding to the gestures. After the gesture recognition system obtains the standard hand data, the gesture recognition system obtains a corresponding standard gesture model from the gesture model database.
S40, determining sample data of the standard gesture model.
Specifically, after the gesture recognition system acquires a certain standard gesture model from the gesture model database, sample data corresponding to the standard gesture model is calculated according to standard hand data in the standard gesture model, wherein the sample data is bending angles of different fingers and palms in the standard gesture model.
S50, comparing the standard hand data with the sample data, and calculating a data difference value between the standard hand data and the sample data.
Specifically, in the process of comparing standard hand data with sample data, the gesture recognition system uses an algorithm to judge the data error between the compared standard hand data and sample data so as to ensure the accuracy of comparison. It should be noted that, according to hand data of hand gesture, a plurality of standard finger data can be obtained by calculation, for example, four key points are shared on the middle finger, so that two standard hand data can be obtained by calculation, when data comparison is performed, different standard hand data need to be compared with corresponding sample data, and data difference values of the two standard hand data need to be calculated, and the number of groups to be compared is not unique, so that the number of obtained data difference values is also not unique.
S60, judging whether the data difference value is within a preset reasonable difference value range.
The gesture recognition system compares the calculated data difference value with a preset endpoint value of a reasonable difference value range, specifically, the gesture recognition system firstly compares the data difference value with a minimum value in the reasonable difference value range, and if the data difference value is smaller than the minimum value, the gesture recognition system directly judges that the standard hand data and the sample data are not successfully matched; if the data difference is greater than the minimum value, the gesture recognition system compares the standard hand data with the maximum value in the reasonable difference range, and if the data difference is greater than the maximum value, the gesture recognition system judges that the standard hand data is not successfully matched with the sample year data, because the data difference is not in the reasonable difference range.
And S70, if the hand data is positioned, judging that the standard hand data is successfully matched with the sample data.
Specifically, if the data difference between the standard hand data and the sample data is greater than the minimum value in the reasonable difference range and less than the maximum value in the reasonable difference range, it can be stated that the standard hand data is located in the reasonable difference range, it can be stated that the hand gesture of the user at the moment is the standard gesture model corresponding to the sample data, that is, the standard hand data and the sample data are successfully matched, and only if all the data differences in the data differences are located in the reasonable difference range, it can be determined that the current hand gesture of the user is the standard hand model currently invoked by the gesture recognition system.
Because of slight differences in shape, size and joint structure of each hand, the standard hand data are difficult to be one hundred percent identical to the standard gesture models stored in the gesture model database in advance, and the standard hand data and the sample data are reasonably controlled, so that the success rate of matching is improved under the condition of ensuring the correct matching result; because the hand data that the hand data acquisition device gathered is mostly finger joint information, in the in-process that the user put out the gesture, the hand structure that plays main role to the hand gesture just is the joint point of finger, consequently, the joint information that gathers the finger is used for gesture recognition can be when reducing the contrast data volume, guarantees gesture recognition result's accuracy.
In one embodiment, to improve the efficiency of gesture recognition, after determining that the standard hand data and the sample data are successfully matched if the gesture is located, the following steps may be further performed:
recording the current gesture model matching according to the matching result when the gesture recognition system completes the gesture model matching every time, specifically, if the matching result of the current gesture model is successful, updating the matching success times corresponding to the standard gesture model in the matching process and the time information when the matching is successful in a gesture model database by the gesture recognition system; when a subsequent gesture recognition system invokes standard gesture models from a preset gesture model database, the gesture recognition system can arrange the standard gesture models in the gesture model database in an ascending order according to the successful matching times corresponding to each standard gesture model, the unique serial numbers corresponding to each standard gesture model are smaller in the serial number values corresponding to the standard gesture models with more successful matching times, and then the gesture recognition system sequentially acquires the standard gesture models according to the serial numbers. According to the gesture recognition method and device, the gesture can be described as the gesture commonly used by the user from the side surface according to the number of successful matching times corresponding to the standard gesture model, and the standard gesture model with higher use frequency is compared preferentially when the gesture recognition is performed, so that the efficiency of the gesture recognition result is improved under the condition that the accuracy of the gesture recognition result is not affected.
In one embodiment, in order to further improve the efficiency of gesture recognition of the gesture recognition system, in the step of arranging the standard gesture models in the gesture model database in ascending order according to the number of successful matches corresponding to the standard gesture model, the following steps may be further executed:
in the process of arranging the standard gesture models in ascending order according to the successful matching times corresponding to the standard gesture models, if the matching success times of a plurality of standard gesture models are equal, the gesture recognition system can query the historical time information when the standard gesture data are successfully matched last time, then order the standard gesture data according to the sequence of the queried historical time information, for the sake of understanding, for example, the three successful matching times corresponding to the three standard gesture models are five times in the current gesture model database, the three standard gesture models are respectively a gesture model A, a gesture model B and a gesture model C, when the gesture recognition system orders the third standard gesture model, firstly, the time data of the three standard gesture models successfully matched last time from the current time node is queried, the time data is the historical time information mentioned above, the historical time information corresponding to the gesture model A is 2023.02.21, the historical time information corresponding to the gesture model B is 2023.01.16, the historical time information corresponding to the gesture model C is 2023.04.15, and the sequence number of the gesture model C is earlier than the sequence number of the gesture model A according to the sequence of the three historical time information. Because the common gestures of the user may have differences in time periods, that is, the common gestures of the user may change in the last time, the standard gesture models are ordered according to the time sequence when the standard gesture data are successfully matched last time, so that the time difference of the common gestures of the user can be considered when the gesture recognition system performs gesture recognition, and the efficiency of gesture recognition is further improved.
In one embodiment, in order to increase the richness of the standard gesture model of the gesture model database, after determining whether the data difference is within the preset reasonable difference range, the following steps may be further performed:
if a standard gesture model which can be successfully matched with the hand gesture placed by the user is not queried in the gesture model database, the gesture is not input in the gesture model database, the gesture recognition system immediately generates a name confirmation instruction and sends the name confirmation instruction to the user terminal, and the user can input the name named for the gesture by the user through an application program interface positioned on the user terminal; secondly, an application program matched with the hand information acquisition device can be used for installing the application degree on the intelligent equipment owned by the user, and the user can use the application program to name a standard gesture model and associate the standard gesture model with the execution operation; and thirdly, the control center of the gesture recognition system is provided with a data communication function among the control center, the hand data acquisition device and the application program, and the control center receives the hand data sent by the hand data acquisition device and performs gesture recognition according to the received hand data.
The application program of the user terminal packages the acquired gesture names to generate name feedback instructions, the name feedback instructions are sent to the control center of the gesture recognition system, then after the gesture models are named, the gesture recognition system obtains initial gesture models, then a user can establish mapping relations between the initial gesture model names and the operation instructions through operation on the application program, the mapping relations are sent to the control center of the gesture recognition system, the control center stores the mapping relations and the initial gesture models to form standard gesture models, the gesture recognition system stores the standard gesture models into the gesture model database, and in the using process of the user, the gesture recognition system can continuously update the gesture model database, so that the richness of the standard gesture models of the gesture model database is greatly improved.
As shown in fig. 2, in one embodiment, to reduce the influence of the fault of the gesture recognition system on the gesture recognition result, if the fault is not located, a name confirmation instruction is sent to the user terminal, which may specifically be executed as the following steps:
when the standard hand data and the sample data are firstly matched and fail, the gesture recognition system performs the matching of the standard hand data and the sample data again, specifically, the gesture recognition system recalculates the data difference between the standard hand data and the sample data, then compares the recalculated data difference with the endpoint value of the reasonable difference range, judges whether the recalculated data difference is in the reasonable difference range, if the matching result between the standard hand data and the sample data is continuously failed, and the failure times reach the preset matching times threshold value, the gesture recognition system sends a name confirmation instruction to the user terminal. When the gesture recognition system calculates the data difference between the standard hand data and the sample data, the possibility of error calculation of the data difference caused by the fault of the gesture recognition system exists, and the gesture recognition system carries out gesture recognition again under the condition of gesture recognition failure, so that the occurrence of the conditions is reduced, and the influence of the fault of the gesture recognition system on the gesture recognition result is reduced.
FIG. 1 is a flow chart of a gesture recognition method in one embodiment. It should be understood that, although the steps in the flowchart of fig. 1 are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows; the steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders; and at least some of the steps in fig. 1 may include a plurality of sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the sub-steps or stages are performed necessarily occur in sequence, but may be performed alternately or alternately with at least some of the other steps or sub-steps of other steps.
Based on the method, the embodiment of the application also discloses a gesture recognition device.
As shown in fig. 3, the apparatus comprises the following modules:
the hand data acquisition module 301 is configured to control the hand data acquisition device to acquire hand data of a user;
the hand data processing module 302 is configured to process a data format of the hand data to obtain standard hand data;
the gesture model retrieving module 303 is configured to retrieve a standard gesture model from a preset gesture model database;
a sample data determination module 304 for determining sample data of the standard gesture model,
the data difference calculation module 305 is configured to compare the standard hand data with the sample data, and calculate a data difference between the standard hand data and the sample data;
a data difference comparison module 306, configured to determine whether the data difference is within a preset reasonable difference range;
the gesture model matching module 307 is configured to determine that the standard hand data and the sample data are successfully matched if the standard hand data are located.
In one embodiment, the hand data acquisition module 301 is further configured to perform hand data, and at least includes: the key point coordinate values and the key point connection relations.
In one embodiment, the hand data processing module 302 is further configured to determine a key point connection relationship in the hand data;
acquiring coordinate values of adjacent key points according to the key point connection relation,
calculating a hand bending angle according to the coordinate values of the key points;
the hand bending angle is recorded as standard hand data.
In one embodiment, the gesture model retrieving module 303 is further configured to record the number of times that each standard gesture model in the gesture model database is successfully matched;
retrieving a standard gesture model from a preset gesture model database, wherein the method specifically comprises the following steps:
according to the successful times of the matching corresponding to the standard gesture models, the standard gesture models in the gesture model database are arranged in ascending order, and each standard gesture model corresponds to a unique serial number;
and sequentially acquiring the standard gesture models according to the serial numbers.
In one embodiment, the gesture model retrieving module 303 is further configured to query, in the process of ascending the standard gesture models, historical time information where the standard gesture data was successfully matched last time if there are a plurality of standard gesture models with equal successful times of matching;
and sequencing the standard gesture data according to the sequence of the historical time information.
In one embodiment, the gesture model retrieving module 303 is further configured to send a name confirmation instruction to the user terminal if the gesture model is not located;
receiving a name feedback instruction sent by a user terminal, wherein the name feedback instruction comprises a gesture model name;
establishing a relationship between the gesture model name and standard hand data, and generating a standard gesture model;
the standard gesture model is stored in a gesture model database.
In one embodiment, the gesture model retrieving module 303 is further configured to match the standard hand data with the sample data again if the standard hand data is not located;
and when the continuous failure times of the standard hand data and the sample data reach a preset matching times threshold, sending a name confirmation instruction to the user terminal.
The embodiment of the application also discloses a computer device.
In particular, the computer device comprises a memory and a processor, the memory having stored thereon a computer program that can be loaded by the processor and that performs the gesture recognition method described above.
The embodiment of the application also discloses a computer readable storage medium.
Specifically, the computer-readable storage medium stores a computer program that can be loaded by a processor and that performs the gesture recognition method as described above, for example, the computer-readable storage medium includes: a U-disk, a removable hard disk, a Read-only memory (ROM), a random access memory (RandomAccessMemory, RAM), a magnetic disk, an optical disk, or other various media capable of storing program codes.
The present embodiment is only for explanation of the present invention and is not to be construed as limiting the present invention, and modifications to the present embodiment, which may not creatively contribute to the present invention as required by those skilled in the art after reading the present specification, are all protected by patent laws within the scope of claims of the present invention.
Claims (6)
1. A method of gesture recognition, the method being applied to a gesture recognition system comprising a hand data acquisition device for acquiring hand data, the method comprising:
controlling a hand data acquisition device to acquire hand data of a user;
processing the hand data to obtain standard hand data;
invoking a standard gesture model from a preset gesture model database;
sample data of the standard gesture model is determined,
comparing the standard hand data with the sample data, and calculating a data difference value between the standard hand data and the sample data;
judging whether the data difference value is within a preset reasonable difference value range or not;
if the standard hand data is located, judging that the standard hand data is successfully matched with the sample data;
after the standard hand data is judged to be successfully matched with the sample data if the standard hand data is located, the method further comprises the following steps:
recording the successful times of matching each standard gesture model in the gesture model database;
the step of retrieving the standard gesture model from a preset gesture model database specifically comprises the following steps:
according to the successful times of the matching corresponding to the standard gesture models, the standard gesture models in the gesture model database are arranged in ascending order, and each standard gesture model corresponds to a unique serial number;
sequentially acquiring standard gesture models according to the serial numbers;
and performing ascending arrangement on the standard gesture models in the gesture model database according to the matching success times corresponding to the standard gesture models, and further comprising:
in the process of ascending order of the standard gesture models, if the matching success times of a plurality of standard gesture models are equal, inquiring historical time information of the last successful matching of the standard gesture models;
sequencing the standard gesture models according to the sequence of the historical time information;
after the judging whether the data difference value is within the preset reasonable difference value range, the method further comprises the following steps:
if not, a name confirmation instruction is sent to the user terminal;
receiving a name feedback instruction sent by a user terminal, wherein the name feedback instruction comprises a gesture model name;
establishing a correlation between the gesture model name and the standard hand data to generate a standard gesture model;
storing the standard gesture model into a gesture model database;
if not, a name confirmation instruction is sent to the user terminal, which specifically comprises:
if not, the standard hand data is matched with the sample data again;
and when the continuous failure times of the standard hand data and the sample data reach a preset matching times threshold, sending a name confirmation instruction to the user terminal.
2. The method of claim 1, wherein the hand data comprises at least: the key point coordinate values and the key point connection relations.
3. The method according to claim 2, wherein the processing the hand data to obtain standard hand data specifically comprises:
determining a key point connection relation in the hand data;
acquiring adjacent key point coordinate values according to the key point connection relation,
calculating a hand bending angle according to the coordinate values of the key points;
and marking the hand bending angle as standard hand data.
4. A gesture recognition apparatus, the apparatus comprising:
the hand data acquisition module (301) is used for controlling the hand data acquisition device to acquire hand data of a user;
the hand data processing module (302) is used for processing the data format of the hand data to obtain standard hand data;
the gesture model retrieving module (303) is used for retrieving a standard gesture model from a preset gesture model database; the method is also used for recording the successful times of matching each standard gesture model in the gesture model database; retrieving a standard gesture model from a preset gesture model database, wherein the method specifically comprises the following steps: according to the successful times of the matching corresponding to the standard gesture models, the standard gesture models in the gesture model database are arranged in ascending order, and each standard gesture model corresponds to a unique serial number; sequentially acquiring standard gesture models according to the serial numbers; the method is also used for inquiring historical time information of the last successful matching of the standard gesture models if the matching success times of the plurality of standard gesture models are equal in the process of ascending order arrangement of the standard gesture models; sorting the standard gesture models according to the sequence of the historical time information; if not, the method is also used for sending a name confirmation instruction to the user terminal; receiving a name feedback instruction sent by a user terminal, wherein the name feedback instruction comprises a gesture model name; establishing a relationship between the gesture model name and standard hand data, and generating a standard gesture model; storing the standard gesture model into a gesture model database; if not, the standard hand data is matched with the sample data again; when the continuous failure times of matching the standard hand data and the sample data reach a preset matching times threshold value, a name confirmation instruction is sent to the user terminal;
a sample data determination module (304) for determining sample data of the standard gesture model,
a data difference calculation module (305) for comparing the standard hand data with the sample data, and calculating a data difference between the standard hand data and the sample data;
a data difference comparison module (306) for judging whether the data difference is within a preset reasonable difference range;
and the gesture model matching module (307) is used for judging that the standard hand data and the sample data are successfully matched if the standard hand data are located.
5. A computer device comprising a memory and a processor, the memory having stored thereon a computer program capable of being loaded by the processor and performing the method according to any of claims 1 to 3.
6. A computer readable storage medium, characterized in that a computer program is stored which can be loaded by a processor and which performs the method according to any of claims 1 to 3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310553653.1A CN116597473B (en) | 2023-05-16 | 2023-05-16 | Gesture recognition method, device, equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310553653.1A CN116597473B (en) | 2023-05-16 | 2023-05-16 | Gesture recognition method, device, equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116597473A CN116597473A (en) | 2023-08-15 |
| CN116597473B true CN116597473B (en) | 2024-01-26 |
Family
ID=87605766
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310553653.1A Active CN116597473B (en) | 2023-05-16 | 2023-05-16 | Gesture recognition method, device, equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116597473B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118097793B (en) * | 2024-04-18 | 2024-07-02 | 广州炫视智能科技有限公司 | Self-adaptive interface gesture operation control system and control method thereof |
| CN118519528A (en) * | 2024-05-29 | 2024-08-20 | 中国标准化研究院 | Interactive control system and method based on gestures |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111126339A (en) * | 2019-12-31 | 2020-05-08 | 北京奇艺世纪科技有限公司 | Gesture recognition method and device, computer equipment and storage medium |
| CN112904994A (en) * | 2019-11-19 | 2021-06-04 | 深圳岱仕科技有限公司 | Gesture recognition method and device, computer equipment and storage medium |
| CN113591519A (en) * | 2020-04-30 | 2021-11-02 | 上海风语筑文化科技股份有限公司 | Gesture recognition processing method |
| CN114792443A (en) * | 2022-04-25 | 2022-07-26 | 扬州大学 | Intelligent device gesture recognition control method based on image recognition |
| WO2022166243A1 (en) * | 2021-02-07 | 2022-08-11 | 青岛小鸟看看科技有限公司 | Method, apparatus and system for detecting and identifying pinching gesture |
| CN115240237A (en) * | 2022-07-26 | 2022-10-25 | 华盛通(无锡)影像科技有限公司 | Multimode-fused finger vein identification method |
| CN115410274A (en) * | 2022-08-30 | 2022-11-29 | 上海高德威智能交通系统有限公司 | Gesture recognition method and device and storage medium |
-
2023
- 2023-05-16 CN CN202310553653.1A patent/CN116597473B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112904994A (en) * | 2019-11-19 | 2021-06-04 | 深圳岱仕科技有限公司 | Gesture recognition method and device, computer equipment and storage medium |
| CN111126339A (en) * | 2019-12-31 | 2020-05-08 | 北京奇艺世纪科技有限公司 | Gesture recognition method and device, computer equipment and storage medium |
| CN113591519A (en) * | 2020-04-30 | 2021-11-02 | 上海风语筑文化科技股份有限公司 | Gesture recognition processing method |
| WO2022166243A1 (en) * | 2021-02-07 | 2022-08-11 | 青岛小鸟看看科技有限公司 | Method, apparatus and system for detecting and identifying pinching gesture |
| CN114792443A (en) * | 2022-04-25 | 2022-07-26 | 扬州大学 | Intelligent device gesture recognition control method based on image recognition |
| CN115240237A (en) * | 2022-07-26 | 2022-10-25 | 华盛通(无锡)影像科技有限公司 | Multimode-fused finger vein identification method |
| CN115410274A (en) * | 2022-08-30 | 2022-11-29 | 上海高德威智能交通系统有限公司 | Gesture recognition method and device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116597473A (en) | 2023-08-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116597473B (en) | Gesture recognition method, device, equipment and storage medium | |
| US20190251471A1 (en) | Machine learning device | |
| JP4041496B2 (en) | Method and apparatus for scaling handwriting input for handwriting recognition | |
| CN101149804B (en) | Self-adaptive hand-written discrimination system and method | |
| CN104866465A (en) | Sensitive text detection method and device | |
| WO2021017286A1 (en) | Facial recognition method and apparatus, electronic device and non-volatile computer readable storage medium | |
| WO2022222943A1 (en) | Department recommendation method and apparatus, electronic device and storage medium | |
| CN111401219B (en) | Palm key point detection method and device | |
| JPWO2020026643A1 (en) | Information processing equipment, information processing methods and information processing programs | |
| US20230368028A1 (en) | Automated machine learning pre-trained model selector | |
| CN111785385A (en) | Disease classification method, device, equipment and storage medium | |
| WO2024078088A1 (en) | Interaction processing method and apparatus | |
| WO2020140624A1 (en) | Method for extracting data from log, and related device | |
| JP4194680B2 (en) | Data processing apparatus and method, and storage medium storing the program | |
| CN116627771B (en) | Log acquisition method, device, electronic equipment and readable storage medium | |
| WO2021047314A1 (en) | Login authentication method and apparatus, and computer device and storage medium | |
| CN111696157B (en) | Image repositioning determination method, system, device and storage medium | |
| CN111353346A (en) | Action recognition method, device, system, electronic equipment and storage medium | |
| CN113743448A (en) | Model training data acquisition method, model training method and device | |
| CN113836005A (en) | Virtual user generation method and device, electronic equipment and storage medium | |
| CN111752654A (en) | Interface layout acquisition method and device, mobile device and storage medium | |
| CN111209407B (en) | Data processing method, device, electronic equipment and computer readable storage medium | |
| CN114942718A (en) | Virtual key setting method, device, medium and electronic equipment | |
| CN114185937B (en) | Big data tracing method and system based on digital finance | |
| CN112068700B (en) | Text information input method, device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |