CN110378169B - Gesture interval detection method and device - Google Patents
Gesture interval detection method and device Download PDFInfo
- Publication number
- CN110378169B CN110378169B CN201810326692.7A CN201810326692A CN110378169B CN 110378169 B CN110378169 B CN 110378169B CN 201810326692 A CN201810326692 A CN 201810326692A CN 110378169 B CN110378169 B CN 110378169B
- Authority
- CN
- China
- Prior art keywords
- interval
- data
- gesture
- preset
- data interval
- 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
Images
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
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The embodiment of the application discloses a method and a device for detecting a gesture interval. The method comprises the steps of acquiring a plurality of data intervals comprising a first preset number of acceleration data acquired in real time, and then obtaining a first preset number of converted acceleration data by using a hyperbolic tangent function for each first acceleration data of the first data intervals, wherein the first data intervals are any data intervals in the plurality of data intervals; obtaining the dispersion of a first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data; and determining whether the first data interval is a gesture interval or not based on the comparison result of the dispersion of the first data interval and a preset dispersion threshold. The method improves the gesture detection interval property, can accurately distinguish the gesture interval from the non-gesture interval in real time, can provide the starting point and the ending point of the gesture signal, and improves the user experience.
Description
Technical Field
The present application relates to the field of gesture recognition, and in particular, to a method and an apparatus for detecting a gesture interval.
Background
Currently, the gesture signal acquisition mode in the gesture recognition field may include the following two modes:
in the first mode, before a user makes a gesture, the user needs to manually turn on a switch of the gesture collection device and then makes a gesture action, and after the gesture action is completed, the user needs to manually turn off the switch of the device, so that collection of a complete gesture data is completed.
However, the whole gesture signal acquisition process in this way needs to be completed through manual operation of the user, and the user experience is poor.
In the second mode, the gesture signals are acquired by comparing the relevant data acquired by the acceleration sensor, the angular velocity sensor and the magnetometer with the set corresponding thresholds respectively and distinguishing the gesture signals from the non-gesture signals based on the comparison result.
However, in this method, when the force is small when the user performs a gesture, the collected related data may be smaller than the preset threshold, and the gesture signal cannot be effectively detected, and for a situation that there is a slight pause in the middle of the user performing a gesture, the gesture motions before and after the pause are likely to be determined as two gesture signals, that is, the collected related data before and after the pause are both larger than the preset threshold, and the collected related data during the pause are smaller than the preset threshold, which results in inaccurate detected gesture signals.
Disclosure of Invention
According to the method and the device for detecting the gesture interval, accuracy of detecting the gesture interval is improved, and therefore user experience is improved.
In a first aspect, a method for detecting a gesture interval is provided, and the method may include:
acquiring a plurality of data intervals, wherein the data intervals comprise a first preset number of acceleration data acquired in real time;
obtaining a first preset number of transformed acceleration data by adopting a hyperbolic tangent function for each first acceleration data of a first data interval, wherein the first data interval is any one of a plurality of data intervals;
obtaining the dispersion of a first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data;
and determining whether the first data interval is a gesture interval or not based on the comparison result of the dispersion of the first data interval and a preset dispersion threshold value, wherein the gesture interval is a data interval belonging to a preset gesture signal.
Therefore, in order to improve the generalization capability of the detection method, namely, inhibit non-gesture data and increase gesture data, the method expands the data range of the first acceleration data through nonlinear transformation, and can accurately distinguish a gesture interval from a non-gesture interval in real time by calculating the dispersion, so that the user experience is improved.
In an alternative implementation, the transformed acceleration data comprises velocity data for a second predetermined number of spatial dimensions;
the acceleration data after the first predetermined number of transformations adopt and predetermine discrete algorithm, obtain the dispersion of first data interval, include:
obtaining speed data of a second preset number of spatial dimensions of the converted acceleration data by adopting a first preset algorithm;
and carrying out variance operation on the first preset number of rate data to obtain the dispersion of the first data interval.
The calculation mode is a calculation mode for calculating the dispersion of the data intervals, and the gesture intervals and the non-gesture intervals of the data intervals are detected by acquiring the difference degree of the first acceleration data in the data intervals reflected by the dispersion.
In an optional implementation, determining whether the first data interval is a gesture interval based on a comparison result of the discrete value of the first data interval and a preset discrete threshold includes:
when the dispersion of the first data interval is larger than a preset dispersion threshold value, determining that the first data interval is a gesture interval; when the dispersion of a first data interval is not larger than a preset dispersion threshold value, acquiring the correlation between the first data interval and a second data interval, wherein the second data interval is a previous data interval adjacent to the first data interval, and when the first data interval is an initial data interval, the second data interval is a preset data interval with zero correlation;
if the correlation degree is larger than a preset correlation degree threshold value, determining that the first data interval is a gesture interval; and if the correlation degree is not greater than the preset correlation degree threshold value, determining that the first data interval is a non-gesture interval.
The calculation method is a calculation method for calculating the correlation degree of the data intervals, and when the difference degree of the first acceleration data in the data intervals reflected by the obtained dispersion degree is low, in order to realize accurate detection of the data intervals, the gesture intervals and the non-gesture intervals of the data intervals are detected by obtaining the correlation degree of the first data intervals and the second data intervals.
In an alternative implementation, obtaining a correlation value between the first data interval and the second data interval includes:
obtaining the relevance of each dimension of each second acceleration data of the first data interval by adopting a preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the first data interval;
obtaining the relevance of each dimension of the second data interval by adopting a preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the second data interval;
and obtaining the correlation degree of the first data interval and the second data interval by adopting a second preset algorithm on the same dimensionality based on the correlation degree of each dimensionality of the first data interval and the correlation degree of each dimensionality of the second data interval.
In an optional implementation, the second preset algorithm is to calculate a maximum value of absolute values of differences between the correlation degrees of the dimensions in the first data interval and the correlation degrees of the corresponding dimensions in the second data interval;
the second predetermined algorithm formula may be expressed as: r ismax＝max{abs(r1 s-r2 s) }; wherein r ismaxThe correlation degree of the first data interval and the second data interval, max { } represents taking the maximum value, abs () represents taking the absolute value, s is the target dimension, r1Is the degree of correlation in the s dimension, r, of the first data interval2Is the correlation degree in the dimension s in the second data interval. The algorithm is a calculation mode of the second preset algorithm.
In an alternative implementation, before obtaining the first preset number of transformed acceleration data, the method further comprises: obtaining filtered acceleration data by adopting a median filtering algorithm on the acceleration data in the first data interval; and stretching the filtered acceleration data to be within a preset data range based on a normalized stretching algorithm.
Therefore, in order to improve the detection accuracy, the median filtering algorithm is used for performing median filtering on the first acceleration data, and the normalization stretching algorithm is used for stretching the data range of the filtered first acceleration data.
In an optional implementation, acquiring the position of a gesture interval, the position of a non-gesture interval and the number of continuous intervals of the non-gesture interval in a plurality of data intervals;
taking the determined position of the first gesture interval as a starting point of a complete gesture signal;
and determining an end point of the complete gesture signal based on a comparison result of the continuous interval number of the non-gesture intervals and a preset non-gesture interval threshold value.
In order to improve the detection accuracy, the end point of the complete gesture signal is accurately determined by acquiring the number of continuous intervals of the non-gesture intervals, so that the accuracy of the starting point and the end point of the detected gesture signal is improved, and the user experience is improved.
In an optional implementation, determining an end point of the complete gesture signal based on a comparison result of the number of continuous intervals of the non-gesture interval and a preset non-gesture interval threshold includes:
when the number of the continuous intervals is not less than a preset non-gesture interval threshold value, determining a gesture interval adjacent to a first non-gesture interval in the continuous non-gesture intervals corresponding to the number of the continuous intervals as an end point of the complete gesture signal; and when the number of the continuous intervals is smaller than a preset non-gesture interval threshold value, modifying the continuous non-gesture interval corresponding to the number of the continuous intervals into a gesture interval, and taking the determined last gesture interval as an end point of the complete gesture signal. This approach is one achievable approach to determining the end point of a full gesture signal.
In a second aspect, an apparatus for detecting a gesture interval is provided, and the apparatus may include:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a plurality of data intervals, and the data intervals comprise a first preset number of acceleration data acquired in real time;
the computing unit is used for obtaining a first preset number of converted acceleration data by adopting a hyperbolic tangent function for each acceleration data of a first data interval, wherein the first data interval is any one of a plurality of data intervals;
obtaining the dispersion of a first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data;
and the determining unit is used for determining whether the first data interval is a gesture interval or not based on the comparison result of the dispersion of the first data interval and a preset dispersion threshold value, and the gesture interval is a data interval belonging to the gesture signal.
In an alternative implementation, the arithmetic unit is specifically configured to:
obtaining first speed data by adopting a first preset algorithm according to the speed data of a second preset number of spatial dimensions of each transformed acceleration data;
and carrying out variance operation on the first preset number of rate data to obtain the dispersion of the first data interval.
In an optional implementation, the determining unit is specifically configured to: when the dispersion of the first data interval is larger than a preset dispersion threshold value, determining that the first data interval is a gesture interval;
when the dispersion of a first data interval is not larger than a preset dispersion threshold value, acquiring the correlation between the first data interval and a second data interval, wherein the second data interval is a previous data interval adjacent to the first data interval, and when the first data interval is an initial data interval, the second data interval is a preset data interval with zero correlation;
if the correlation degree is larger than a preset correlation degree threshold value, determining that the first data interval is a gesture interval;
and if the correlation degree is not greater than the preset correlation degree threshold value, determining that the first data interval is a non-gesture interval.
In an alternative implementation, the determining unit is further configured to:
obtaining the relevance of each dimension of the first data interval by adopting a preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the first data interval;
obtaining the relevance of each dimension of the second data interval by adopting a preset relevance algorithm on the speed data in the same dimension in a preset number of transformed acceleration data corresponding to the second data interval;
and obtaining the correlation degree of the first data interval and the second data interval by adopting a second preset algorithm on the same dimensionality based on the correlation degree of each dimensionality of the first data interval and the correlation degree of each dimensionality of the second data interval.
In an optional implementation, the second preset algorithm is to calculate a maximum value of absolute values of differences between the correlation degrees of the dimensions in the first data interval and the correlation degrees of the corresponding dimensions in the second data interval; the second predetermined algorithm formula may be expressed as: r ismax＝max{abs(r1 s-r2 s) }; wherein r ismaxThe correlation degree of the first data interval and the second data interval, max { } represents taking the maximum value, abs () represents taking the absolute value, s is the target dimension, r1Is the degree of correlation in the s dimension, r, of the first data interval2Is the correlation degree in the dimension s in the second data interval.
In an optional implementation, the operation unit is further configured to obtain filtered acceleration data by using a median filtering algorithm on the acceleration data in the first data interval before obtaining the first preset number of transformed acceleration data;
and stretching the filtered acceleration data to be within a preset data range based on a normalized stretching algorithm.
In an optional implementation, the obtaining unit is further configured to obtain a position of a gesture interval, a position of a non-gesture interval, and a number of consecutive intervals of the non-gesture interval in the plurality of data intervals;
the determining unit is further used for taking the determined position of the first gesture interval as a starting point of the complete gesture signal;
and determining an end point of the complete gesture signal based on a comparison result of the continuous interval number of the non-gesture intervals and a preset non-gesture interval threshold value.
In an optional implementation, the determining unit is further specifically configured to: when the number of the continuous intervals is not less than a preset non-gesture interval threshold value, determining a gesture interval adjacent to a first non-gesture interval in the continuous non-gesture intervals corresponding to the number of the continuous intervals as an end point of the complete gesture signal;
and when the number of the continuous intervals is smaller than a preset non-gesture interval threshold value, modifying the continuous non-gesture interval corresponding to the number of the continuous intervals into a gesture interval, and taking the determined last gesture interval as an end point of the complete gesture signal.
In a fifth aspect, an electronic device is provided, which includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete communication with each other via the communication bus;
a memory for storing a computer program;
a processor configured to perform the method steps of any one of the first aspect or the method steps of any one of the second aspect when executing the program stored in the memory.
A sixth aspect provides a computer readable storage medium having stored therein a computer program which, when executed by a processor, performs the method steps of any one of the above first aspects or the method steps of any one of the above second aspects.
The method comprises the steps of acquiring a plurality of data intervals comprising a first preset number of acceleration data acquired in real time, and then obtaining a first preset number of converted acceleration data from each first acceleration data of the first data intervals by adopting a hyperbolic tangent function, wherein the first data intervals are any data intervals of the plurality of data intervals; obtaining the dispersion of a first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data; and determining whether the first data interval is a gesture interval or not based on the comparison result of the dispersion of the first data interval and a preset dispersion threshold. The method improves the gesture detection interval property, can accurately distinguish the gesture interval from the non-gesture interval in real time, can provide the starting point and the ending point of the gesture signal, and improves the user experience.
Drawings
Fig. 1 is a schematic flowchart of a gesture interval detection method based on acceleration data according to an embodiment of the present invention;
FIG. 2 is a graphical illustration of the range of a tanh function;
fig. 3A is a schematic distribution diagram of gesture intervals and non-gesture intervals according to an embodiment of the present invention;
FIG. 3B is a schematic diagram illustrating another distribution of gesture intervals and non-gesture intervals according to an embodiment of the present invention;
fig. 4 is a schematic flowchart of a method for detecting a gesture interval according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a detection apparatus for detecting a gesture interval according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present application without any creative effort belong to the protection scope of the present application.
The method for detecting the gesture interval and the gesture endpoint in the acceleration data provided by the embodiment of the invention can be applied to the terminal. The terminal referred to in the present application may include various handheld devices having a wireless communication function, vehicle-mounted devices, Wearable Devices (WD), computing devices or other processing devices connected to a wireless modem, and various forms of User Equipment (UE), Mobile Stations (MS), terminals (terminal), terminal devices (terminal equipment), and the like, for collecting motion data of a user's arm.
The terminal can be divided into a wearable terminal and a high-performance terminal, wherein the wearable terminal can be directly worn on the body, or can also be a portable device integrated with clothes or accessories of the arm of a user, such as a watch, a wrist strap and the like; the high-performance terminal is a terminal with higher performance than a wearable terminal, such as a mobile phone, a notebook computer and other user equipment. Wherein, the wearable terminal can only contain the acceleration sensor.
The detection method of the present application uses data intervals of equal length of acceleration data as a calculation unit, that is, the number of acceleration data included in each data interval is the same. Firstly, the detection method includes the steps of acquiring acceleration data of an acceleration sensor in real time, dividing the acceleration data into non-overlapping data intervals with equal length, and obtaining the dispersion of each data interval after the acceleration data in the data intervals are subjected to nonlinear transformation and a series of calculations, wherein the dispersion of the data intervals is used for representing the distribution condition (or called difference degree) of the acceleration data in the data intervals, and the larger the dispersion is, the larger the difference of the acceleration data in the data intervals is, the higher the probability of the acceleration data belonging to a gesture interval is, and otherwise, the smaller the dispersion is, the higher the probability of the acceleration data belonging to a non-gesture interval is.
And determining whether the data interval is a gesture interval or not based on the comparison result of the dispersion of the data interval and a preset dispersion threshold, wherein the gesture interval is a data interval belonging to a gesture signal, and the gesture signal is a signal of a preset gesture track. Secondly, detecting the end point detection of the complete gesture signal based on the determined gesture interval and the non-gesture interval, namely detecting the starting point and the end point of the complete gesture signal, wherein the complete gesture signal is a signal with a preset complete gesture track.
Therefore, the detection method can detect the gesture interval and the non-gesture interval only based on the acceleration sensor, compared with the sensors in the prior art, the detection cost is low, the endpoint of the complete gesture signal can be accurately determined in the detected gesture interval and the non-gesture interval, and the problem that one gesture signal is divided into two different gesture signals due to short pause in the gesture process is solved. The method can achieve a good detection effect for different types of users, such as users with strength, and improves the generalization capability of the detection method, wherein the generalization capability is the adaptability of the detection method to new users.
The preferred embodiments of the present application will be described below with reference to the accompanying drawings of the specification, it being understood that the preferred embodiments described herein are merely for illustrating and explaining the present invention and are not intended to limit the present invention, and that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
Fig. 1 is a schematic flow chart of a method for detecting a gesture interval according to an embodiment of the present invention, where an execution subject of the method is a detection device, and as shown in fig. 1, the method may include:
and step 110, acquiring a plurality of data intervals.
The detection device of the application uses the acceleration sensor to collect acceleration data, and the detection device is generally worn on the wrist of a user to collect the acceleration data of the arm of the user in real time. The acceleration data may include velocity data for a second preset number of spatial dimensions, e.g., the acceleration data may include velocity data for 3 spatial dimensions of the X-axis, the Y-axis, and the Z-axis.
The difference is large when different users do the same gesture, and if some users do the gesture, the amplitude is small, the speed is slow, and some users do the same gesture, the amplitude is large, the speed is fast, so that the sampling frequency of the acceleration sensor is required to be higher, and enough data can be acquired for some gestures with fast speed. Therefore, the sampling frequency of the acceleration sensor should not be lower than 50Hz, and the sampling frequency of the acceleration sensor in the embodiment of the present invention is preferably 50 Hz.
In order to detect gesture data in real time, the embodiment of the invention divides the acceleration data acquired in real time into data intervals containing a preset number of acceleration data, and detects the acceleration data by taking each data interval as a unit. Therefore, detection is not required to be carried out after a complete gesture is completed, and the real-time performance of detection is greatly improved. In order to improve the detection accuracy and the detection time, the preset number n may be between 5 and 10.
And step 120, obtaining a first preset number of transformed acceleration data by using a hyperbolic tangent function for each acceleration data of the first data interval.
Wherein, the first data interval is any one data interval in a plurality of data intervals.
Optionally, to improve the detection accuracy, before performing this step, a median filtering algorithm may be applied to the acceleration data in the first data interval to obtain filtered acceleration data. And because noise points exist in the acquired acceleration data, smoothing the acquired acceleration data by using median filtering so as to remove noise.
Taking the second preset number of 3 as an example, since the acceleration data may include 3 dimensions of speed data of X-axis, Y-axis and Z-axis, the range of the speed data in each dimension is typically ± g, i.e. the range of the speed data is [ -1,1 ]. In order to improve the detection accuracy, the filtered acceleration data is stretched to within a preset data range, i.e. the range of the velocity data is expanded to [ -2,2], based on a normalized stretching algorithm.
Further, the main reason that the conventional endpoint detection algorithm does not have the generalization capability or has poor generalization capability is that it is difficult to find a suitable threshold of the acceleration data (i.e., the original acceleration data) to implement the generalization capability because the difference between the gesture data and the non-gesture data in the acquired acceleration data is small.
In order to improve the generalization capability of the detection method, namely, inhibit non-gesture data and increase gesture data, each acceleration data in the data interval is subjected to a preset nonlinear transformation algorithm to obtain a preset number of transformed acceleration data in the data interval.
Wherein the hyperbolic tangent function is a tanh function expressed as:
the tanh function may normalize the stretched transformed acceleration data from [ -2,2]Nonlinear transformation to [ -1,1 [)]Thereby making the numerical conversion of the converted acceleration data close to 0 smaller, the numerical value of the converted acceleration data close to ± 2 is close to ± 1 as shown in fig. 2. Is a range diagram of the tanh function.
And step 130, obtaining the dispersion of the first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data.
Obtaining a first preset number of speed data by adopting a first preset algorithm on the speed data of a second preset number of spatial dimensions of each transformed acceleration data; and carrying out variance operation on the first preset number of rate data to obtain the dispersion of the first data interval.
Wherein the first predetermined algorithm may be expressed as squaring the sum of the squares of the velocity data in the dimensions of each transformed acceleration data, e.g. each velocity data v may be expressed as
Wherein X, Y, Z represent velocity data in three different dimensions, respectively.
In this step, the gesture intervals and the non-gesture intervals of the plurality of data intervals are detected by obtaining the difference degree between the converted acceleration data in the data intervals reflected by the dispersion.
Step 140, determining whether the first data interval is a gesture interval or not based on a comparison result of the dispersion of the first data interval and a preset dispersion threshold.
When the dispersion of the first data interval is larger than a preset dispersion threshold value, determining that the first data interval is a gesture interval;
when the dispersion of the first data interval is not larger than a preset dispersion threshold, the correlation between the first data interval and a second data interval is acquired, and the second data interval is a previous data interval adjacent to the first data interval in the plurality of data intervals. The correlation degree refers to the similarity degree between two data intervals, and when the first data interval is the initial data interval, the second data interval is the preset data interval with zero correlation degree.
If the correlation degree is larger than a preset correlation degree threshold value, determining that the first data interval is a gesture interval;
and if the correlation degree is not greater than the preset correlation degree threshold value, determining that the first data interval is a non-gesture interval.
The manner of obtaining the correlation between the first data interval and the second data interval may include:
obtaining the relevance of each dimension of the first data interval by adopting a preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the first data interval;
obtaining the relevance of each dimension of the second data interval by adopting a preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the second data interval;
and obtaining the correlation degree of the first data interval and the second data interval by adopting a second preset algorithm on the same dimensionality based on the correlation degree of each dimensionality of the first data interval and the correlation degree of each dimensionality of the second data interval.
It should be noted that the correlation degree refers to a degree of similarity between two vectors. The preset vector a ═ aiI ∈ [0, M) }, vector V ═ VjJ belongs to [0, N) ], wherein M is larger than or equal to N. The correlation between vector V and vector a is expressed as: r (V, A) { R ═ Rk,k∈[0,M-N+1)}，
Wherein r iskRepresenting the k-th component of the correlation valid data between vector V and vector a.
Since the number of vectors in each dimension in the first data interval is also the same, i.e., M is equal to N, the correlation of the velocity data in each dimension is a single value r0。
Assuming that the dimension of the acceleration data of the sensor is L, the correlation between the dimensions is common
In this case, the correlation of the dimensional data of each acceleration data in one data interval is represented as:
based on the scheme, when the difference degree of the first acceleration data in the data interval reflected by the obtained dispersion is low, in order to accurately detect the data interval, the gesture intervals and the non-gesture intervals of the multiple data intervals are detected by obtaining the correlation degree of the first data interval and the second data interval.
For example, taking the dispersion of the first data interval as a as an example, the determining that the first data interval is a gesture interval or a non-gesture interval may include:
when the dispersion A is larger than a preset dispersion threshold value, judging that the first data interval is a gesture interval;
when the dispersion A is not larger than a preset dispersion threshold value and the correlation is larger than a preset correlation threshold value, judging that the first data interval is a gesture interval;
and when the dispersion A is not greater than a preset dispersion threshold value and the correlation is not greater than a preset correlation threshold value, judging that the first data interval is a non-gesture interval.
Optionally, the second preset algorithm is an algorithm that calculates a maximum value of absolute values of differences between the correlation degrees of the dimensions in the first data interval and the correlation degrees of the corresponding dimensions in the second data interval, and may be represented as: r ismax＝max{abs(r1 s-r2 s) Wherein r ismaxThe correlation degree of the first data interval and the second data interval, max { } represents taking the maximum value, abs () represents taking the absolute value, s is the target dimension, r1Is the degree of correlation in the s dimension, r, of the first data interval2Is the correlation degree in the dimension s in the second data interval.
Further, after the gesture interval and the non-gesture interval determined based on steps 110-140, the embodiment of the present invention may further detect a gesture endpoint, and perform the following steps:
and 150, acquiring the positions of gesture intervals, the positions of non-gesture intervals and the number of continuous intervals of the non-gesture intervals in the plurality of data intervals.
The number of consecutive intervals may take the value 1.
Based on the gesture interval and the non-gesture interval determined in step 110 and 140, the position of the gesture interval, the position of the non-gesture interval, and the number of continuous intervals of the non-gesture interval in the plurality of data intervals are obtained.
And step 160, taking the determined position of the first gesture interval as a starting point of the complete gesture signal.
And 170, determining an end point of the complete gesture signal based on a comparison result of the number of continuous intervals of the non-gesture intervals and a preset non-gesture interval threshold value.
In order to eliminate the condition of pause in the complete gesture signal, when the number of continuous intervals is not less than a preset non-gesture interval threshold value, determining a gesture interval adjacent to a first non-gesture interval in a plurality of non-gesture intervals corresponding to the number of continuous intervals as an end point of the complete gesture signal;
when the number of continuous intervals is less than the preset non-gesture interval threshold,
and modifying the continuous non-gesture interval corresponding to the continuous interval number into a gesture interval, and taking the determined last gesture interval as an end point of the complete gesture signal.
For example, as shown in fig. 3A and 3B, each box represents a data interval, S represents a gesture interval, and N represents a non-gesture interval.
The first case shown in fig. 3A is: in 8 data intervals, detecting that a position 1 and a position 2 are gesture intervals, positions 3-7 are non-gesture intervals, and positions 7 and 8 are gesture intervals, namely 4 non-gesture intervals are in the gesture intervals;
the second case shown in fig. 3B is: in the 5 data intervals, the gesture intervals are detected as position 1 and position 2, the gesture interval is detected as position 3, the gesture interval is detected as position 4, and the gesture interval is detected as position 5, namely the gesture interval has 1 non-gesture interval.
For the two situations, it is necessary to determine whether the middle non-gesture signal interval belongs to a situation of slight pause in the gesture process.
Taking the preset non-gesture interval threshold as 3 as an example, the number of consecutive intervals of the non-gesture interval in fig. 3A is 4, and the number of consecutive intervals of the non-gesture interval in fig. 3B is 1.
Since the number of consecutive intervals of the non-gesture interval in fig. 3A is not less than 3, the gesture interval before position 3 (i.e., position 2) in the consecutive non-gesture interval is used as the end point of the complete gesture signal.
Since the number of continuous intervals of the non-gesture interval in fig. 3B is less than 3, the non-gesture interval is modified into the gesture interval, and the gesture interval at position 5 is used as the end point of the complete gesture signal. Therefore, the situation of short pause in the gesture process can be eliminated, and the problem that one gesture signal is divided into two different gesture signals is solved.
It should be noted that the step numbers in step 130 and step 140 are only used to indicate the number of steps, and there is no specific execution order, that is, step 140 may be executed before step 130.
The following description will be made in detail, taking as an example that the sampling frequency of the acceleration sensor is preferably 50 Hz.
Fig. 4 is a diagram illustrating another method for detecting a gesture endpoint based on acceleration data according to an embodiment of the present invention, as shown in fig. 4, the method may include:
and step 401, acquiring 5 data intervals in 1S in real time.
Each data interval contains 10 pieces of first acceleration data.
Step 402, a median filtering algorithm and a normalized stretching algorithm are respectively adopted for the 10 pieces of first acceleration data of each data interval, so as to obtain 10 pieces of processed first acceleration data of the data interval.
And step 403, obtaining 10 second acceleration data of each data interval from the 10 processed first acceleration data of each data interval by adopting a hyperbolic tangent function algorithm.
And step 404, obtaining the dispersion of the first data interval by adopting a preset dispersion algorithm for the 10 second acceleration data of each data interval.
Step 405, judging whether the dispersion of each data interval is greater than a preset dispersion threshold value;
if yes, go to step 409;
if not, go to step 406.
And 406, obtaining the relevance of each dimension of each second acceleration data in each data interval by using a preset relevance algorithm for the speed data in the same dimension in each second acceleration data in each data interval.
And 407, obtaining the correlation degree between each data interval and the next adjacent data interval by adopting a second preset algorithm on the same dimension based on the correlation degree between each dimension of each data interval and the correlation degree between each dimension of the next adjacent data interval.
And step 408, judging whether the correlation degree is greater than a preset correlation degree threshold value.
If yes, go to step 409;
if not, go to step 410.
And step 410, determining that the interval in each data interval is a non-gesture interval.
Step 411, obtaining the position of the gesture interval, the position of the non-gesture interval and the number of continuous intervals of the non-gesture interval in the 10 data intervals.
And step 412, taking the determined position of the first gesture interval as a starting point of the complete gesture signal.
Step 413, judging whether the number of continuous intervals of the non-gesture intervals is larger than a preset non-gesture interval threshold value;
if yes, go to step 414;
if not, go to step 415.
And 414, determining a gesture interval adjacent to the first non-gesture interval in the plurality of non-gesture intervals corresponding to the continuous interval number as an end point of the complete gesture signal.
Step 415, modifying the continuous non-gesture interval corresponding to the number of the continuous intervals into a gesture interval, and taking the determined last gesture interval as an end point of the complete gesture signal.
The detection method provided by the embodiment of the invention comprises the following steps: acquiring a plurality of data intervals comprising a first preset number of acceleration data acquired in real time, and then obtaining a first preset number of converted acceleration data from each first acceleration data of the first data intervals by adopting a hyperbolic tangent function, wherein the first data intervals are any data intervals in the plurality of data intervals; obtaining the dispersion of a first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data; and determining whether the first data interval is a gesture interval or not based on the comparison result of the dispersion of the first data interval and a preset dispersion threshold. The method improves the gesture detection interval property, can accurately distinguish the gesture interval from the non-gesture interval in real time, can provide the starting point and the ending point of the gesture signal, and improves the user experience.
Corresponding to the foregoing method, an embodiment of the present invention further provides a device for detecting a gesture interval, where as shown in fig. 5, the device may include: an acquisition unit 510, an arithmetic unit 520 and a determination unit 530.
The acquiring unit 510 is configured to acquire a plurality of data intervals, where each data interval includes a first preset number of non-overlapping first acceleration data acquired in real time, and the first acceleration data includes second preset number of dimensions of speed data;
an operation unit 520, configured to obtain a first preset number of transformed acceleration data by using a hyperbolic tangent function on each first acceleration data in a first data interval of a plurality of data intervals, where the first data interval is any one of the plurality of data intervals;
obtaining the dispersion of a first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data;
the determining unit 530 is configured to determine whether the first data interval is a gesture interval based on a comparison result between the dispersion of the first data interval and a preset dispersion threshold, where the gesture interval is a data interval belonging to the gesture signal.
Optionally, the operation unit 520 is specifically configured to: obtaining speed data of a second preset number of spatial dimensions of the converted acceleration data by adopting a first preset algorithm;
and carrying out variance operation on the first preset number of rate data to obtain the dispersion of the first data interval.
Optionally, the determining unit 510 is specifically configured to: when the dispersion of the first data interval is larger than a preset dispersion threshold value, determining that the first data interval is a gesture interval;
when the dispersion of a first data interval is not larger than a preset dispersion threshold value, acquiring the correlation between the first data interval and a second data interval, wherein the second data interval is a previous data interval adjacent to the first data interval in the plurality of data intervals, and when the first data interval is an initial data interval, the second data interval is a preset data interval with zero correlation;
if the correlation degree is larger than a preset correlation degree threshold value, determining that the first data interval is a gesture interval;
and if the correlation degree is not greater than the preset correlation degree threshold value, determining that the first data interval is a non-gesture interval.
Optionally, the determining unit 510 is further configured to: obtaining the relevance of each dimension of the first data interval by adopting a preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the first data interval;
obtaining the relevance of each dimension of the second data interval by adopting a preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the second data interval;
and obtaining the correlation degree of the first data interval and the second data interval by adopting a second preset algorithm on the same dimensionality based on the correlation degree of each dimensionality of the first data interval and the correlation degree of each dimensionality of the second data interval.
Optionally, the second preset algorithm is to calculate a maximum value of absolute values of differences between the correlation degrees of the dimensions in the first data interval and the correlation degrees of the corresponding dimensions in the second data interval; the second predetermined algorithm formula may be expressed as: r ismax＝max{abs(r1 s-r2 s)}；
Wherein r ismaxThe correlation degree of the first data interval and the second data interval, max { } represents taking the maximum value, abs () represents taking the absolute value, s is the target dimension, r1Is the degree of correlation in the s dimension, r, of the first data interval2Is the correlation degree in the dimension s in the second data interval.
Optionally, the operation unit 520 is further configured to, before obtaining a first preset number of transformed acceleration data, obtain filtered first acceleration data by using a median filtering algorithm on the first acceleration data in the first data interval;
and based on a normalized stretching algorithm, stretching the filtered first acceleration data to be within a preset data range.
Optionally, the predetermined nonlinear transformation algorithm is a hyperbolic tangent function algorithm.
Optionally, the obtaining unit 510 is further configured to obtain a position of a gesture interval, a position of a non-gesture interval, and a continuous interval number of the non-gesture interval in the multiple data intervals;
the determining unit 530 is further configured to use the determined position of the first gesture interval as a starting point of the complete gesture signal;
and determining an end point of the complete gesture signal based on a comparison result of the continuous interval number of the non-gesture intervals and a preset non-gesture interval threshold value.
Optionally, the determining unit 530 is further specifically configured to: when the number of the continuous intervals is not less than a preset non-gesture interval threshold value, determining a gesture interval adjacent to a first non-gesture interval in the continuous non-gesture intervals corresponding to the number of the continuous intervals as an end point of the complete gesture signal;
and when the number of the continuous intervals is smaller than a preset non-gesture interval threshold value, modifying the continuous non-gesture interval corresponding to the number of the continuous intervals into a gesture interval, and taking the determined last gesture interval as an end point of the complete gesture signal.
After obtaining a plurality of data intervals including a first preset number of acceleration data collected in real time, each functional unit of the device provided in the above embodiment of the present invention obtains a first preset number of converted acceleration data from each first acceleration data of the first data interval by using a hyperbolic tangent function, where the first data interval is any one of the plurality of data intervals; obtaining the dispersion of a first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data; and determining whether the first data interval is a gesture interval or not based on the comparison result of the dispersion of the first data interval and a preset dispersion threshold. The method improves the gesture detection interval property, can accurately distinguish the gesture interval from the non-gesture interval in real time, can provide the starting point and the ending point of the gesture signal, and improves the user experience.
An embodiment of the present invention further provides an electronic device, as shown in fig. 6, including a processor 610, a communication interface 620, a memory 630, and a communication bus 640, where the processor 610, the communication interface 620, and the memory 630 complete mutual communication through the communication bus 640.
A memory 630 for storing computer programs;
the processor 610, when executing the program stored in the memory 630, implements the following steps:
acquiring a plurality of data intervals, wherein the data intervals comprise a first preset number of acceleration data acquired in real time;
obtaining a first preset number of transformed acceleration data by adopting a hyperbolic tangent function for each first acceleration data of a first data interval, wherein the first data interval is any one of a plurality of data intervals;
obtaining the dispersion of a first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data;
and determining whether the first data interval is a gesture interval or not based on the comparison result of the dispersion of the first data interval and a preset dispersion threshold, wherein the gesture interval is a data interval belonging to the gesture signal.
As the implementation and the beneficial effects of the problem solving of each device of the apparatus in the foregoing embodiment can be realized by referring to each step in the embodiment shown in fig. 1, detailed working processes and beneficial effects of the apparatus provided in the embodiment of the present invention are not described herein again.
The aforementioned communication bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown, but this does not mean that there is only one bus or one type of bus.
The communication interface is used for communication between the electronic equipment and other equipment.
The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; but also Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components.
In another embodiment of the present invention, a computer-readable storage medium is further provided, where instructions are stored, and when the instructions are executed on a computer, the computer is caused to execute the method for detecting a gesture interval in any one of the above embodiments.
In another embodiment of the present invention, there is also provided a computer program product containing instructions, which when run on a computer, causes the computer to execute the method for detecting a gesture interval described in any of the above embodiments.
As will be appreciated by one of skill in the art, the embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, embodiments of the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Embodiments of the present application are described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including the preferred embodiment and all changes and modifications that fall within the true scope of the embodiments of the present application.
It is apparent that those skilled in the art can make various changes and modifications to the embodiments of the present application without departing from the spirit and scope of the embodiments of the present application. Thus, if such modifications and variations of the embodiments of the present application fall within the scope of the claims of the embodiments of the present application and their equivalents, the embodiments of the present application are also intended to include such modifications and variations.
Claims (16)
1. A method for detecting a gesture interval, the method comprising:
acquiring a plurality of data intervals, wherein the data intervals comprise a first preset number of acceleration data acquired in real time;
obtaining a first preset number of transformed acceleration data by adopting a hyperbolic tangent function for each acceleration data of a first data interval, wherein the first data interval is any one of the plurality of data intervals;
obtaining the dispersion of the first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data;
determining whether the first data interval is a gesture interval or not based on a comparison result of the dispersion of the first data interval and a preset dispersion threshold, wherein the gesture interval is a data interval belonging to a gesture signal;
wherein, based on the comparison result of the discrete value of the first data interval and the preset discrete threshold, determining whether the first data interval is a gesture interval, including:
when the dispersion of the first data interval is larger than a preset dispersion threshold value, determining that the first data interval is a gesture interval;
when the dispersion of the first data interval is not larger than the preset dispersion threshold, acquiring the correlation between the first data interval and a second data interval, wherein the second data interval is a previous data interval adjacent to the first data interval, and when the first data interval is an initial data interval, the second data interval is a preset data interval with zero correlation;
if the correlation degree is larger than a preset correlation degree threshold value, determining that the first data interval is a gesture interval;
and if the correlation degree is not greater than a preset correlation degree threshold value, determining that the first data interval is a non-gesture interval.
2. The method of claim 1, wherein the transformed acceleration data comprises velocity data for a second preset number of spatial dimensions;
obtaining the dispersion of the first data interval by adopting a preset dispersion algorithm on the converted acceleration data of the first preset number, wherein the dispersion comprises the following steps:
obtaining speed data of a second preset number of spatial dimensions of the converted acceleration data by adopting a first preset algorithm;
and carrying out variance operation on the first preset number of rate data to obtain the dispersion of the first data interval.
3. The method of claim 1, wherein obtaining the correlation of the first data interval and the second data interval comprises:
obtaining the relevance of each dimension of the first data interval by adopting a preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the first data interval;
obtaining the relevance of each dimension of the second data interval by adopting the preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the second data interval;
and obtaining the correlation degree of the first data interval and the second data interval by adopting a second preset algorithm on the same dimension based on the correlation degree of each dimension of the first data interval and the correlation degree of each dimension of the second data interval.
4. The method of claim 3, wherein the second predetermined algorithm is to calculate a maximum value of absolute differences between the correlation in each dimension of the first data interval and the correlation in the corresponding dimension of the second data interval;
the second predetermined algorithm formula may be expressed as:
wherein r ismaxFor the correlation degree of the first data interval and the second data interval, max { } represents taking the maximum value, abs () represents taking the absolute value, s is the target dimension, r1Is the degree of correlation in the dimension s, r, of the first data interval2And the correlation degree in the dimension s in the second data interval.
5. The method of claim 1, wherein prior to obtaining the first predetermined number of transformed acceleration data, the method further comprises:
obtaining filtered acceleration data by adopting a median filtering algorithm on the acceleration data in the first data interval;
and stretching the filtered acceleration data to be within a preset data range based on a normalized stretching algorithm.
6. The method of claim 1, wherein the method further comprises:
acquiring the position of a gesture interval, the position of a non-gesture interval and the number of continuous intervals of the non-gesture interval in the plurality of data intervals;
taking the determined position of the first gesture interval as a starting point of a complete gesture signal;
and determining an end point of the complete gesture signal based on a comparison result of the number of continuous intervals of the non-gesture intervals and a preset non-gesture interval threshold value.
7. The method of claim 6, wherein determining an end point of the full gesture signal based on a comparison of a number of consecutive intervals of the non-gesture interval to a preset non-gesture interval threshold comprises:
when the number of the continuous intervals is not smaller than a preset non-gesture interval threshold value, determining that a previous gesture interval adjacent to a first non-gesture interval in the continuous non-gesture intervals corresponding to the number of the continuous intervals is an end point of the complete gesture signal;
and when the number of the continuous intervals is smaller than a preset non-gesture interval threshold value, modifying the continuous non-gesture interval corresponding to the number of the continuous intervals into a gesture interval, and taking the determined last gesture interval as an end point of the complete gesture signal.
8. An apparatus for detecting gesture intervals, the apparatus comprising:
the device comprises an acquisition unit, a processing unit and a processing unit, wherein the acquisition unit is used for acquiring a plurality of data intervals, and the data intervals comprise a first preset number of acceleration data acquired in real time;
the computing unit is used for obtaining a first preset number of converted acceleration data by adopting a hyperbolic tangent function for each acceleration data of a first data interval, wherein the first data interval is any one of a plurality of data intervals;
obtaining the dispersion of the first data interval by adopting a preset dispersion algorithm on the first preset number of converted acceleration data;
the determining unit is used for determining whether the first data interval is a gesture interval or not based on a comparison result of the dispersion of the first data interval and a preset dispersion threshold value, wherein the gesture interval is a data interval belonging to a gesture signal;
the determining unit is specifically configured to:
when the dispersion of the first data interval is larger than a preset dispersion threshold value, determining that the first data interval is a gesture interval;
when the dispersion of the first data interval is not larger than a preset dispersion threshold value, acquiring the correlation between the first data interval and a second data interval, wherein the second data interval is a previous data interval adjacent to the first data interval, and when the first data interval is an initial data interval, the second data interval is a preset data interval with zero correlation;
if the correlation degree is larger than a preset correlation degree threshold value, determining that the first data interval is a gesture interval;
and if the correlation degree is not greater than a preset correlation degree threshold value, determining that the first data interval is a non-gesture interval.
9. The apparatus of claim 8, wherein the arithmetic unit is specifically configured to:
obtaining speed data of a second preset number of spatial dimensions of the converted acceleration data by adopting a first preset algorithm;
and carrying out variance operation on the first preset number of rate data to obtain the dispersion of the first data interval.
10. The apparatus of claim 8, wherein the determining unit is further configured to:
obtaining the relevance of each dimension of the first data interval by adopting a preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the first data interval;
obtaining the relevance of each dimension of the second data interval by adopting the preset relevance algorithm on the speed data of the same dimension in each transformed acceleration data corresponding to the second data interval;
and obtaining the correlation degree of the first data interval and the second data interval by adopting a second preset algorithm on the same dimension based on the correlation degree of each dimension of the first data interval and the correlation degree of each dimension of the second data interval.
11. The apparatus of claim 10, wherein the second predetermined algorithm is to calculate a maximum value of absolute differences between the correlation in each dimension of the first data interval and the correlation in the corresponding dimension of the second data interval;
the second predetermined algorithm formula may be expressed as:
wherein r ismaxFor the correlation degree of the first data interval and the second data interval, max { } represents taking the maximum value, abs () represents taking the absolute value, s is the target dimension, r1Is the degree of correlation in the dimension s, r, of the first data interval2And the correlation degree in the dimension s in the second data interval.
12. The apparatus of claim 8,
the operation unit is further configured to obtain filtered acceleration data by using a median filtering algorithm on the acceleration data in the first data interval before obtaining a first preset number of transformed acceleration data;
and stretching the filtered acceleration data to be within a preset data range based on a normalized stretching algorithm.
13. The apparatus of claim 8,
the acquiring unit is further configured to acquire a position of a gesture interval, a position of a non-gesture interval, and a continuous interval number of the non-gesture interval in the plurality of data intervals;
the determining unit is further configured to use the determined position of the first gesture interval as a starting point of the complete gesture signal;
and determining an end point of the complete gesture signal based on a comparison result of the number of continuous intervals of the non-gesture intervals and a preset non-gesture interval threshold value.
14. The apparatus of claim 13, wherein the second determining unit is specifically configured to:
when the number of the continuous intervals is not smaller than a preset non-gesture interval threshold value, determining a gesture interval adjacent to a first non-gesture interval in the continuous non-gesture intervals corresponding to the number of the continuous intervals as an end point of the complete gesture signal;
and when the number of the continuous intervals is smaller than a preset non-gesture interval threshold value, modifying the continuous non-gesture interval corresponding to the number of the continuous intervals into a gesture interval, and taking the determined last gesture interval as an end point of the complete gesture signal.
15. An electronic device, characterized in that the electronic device comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory are communicated with each other through the communication bus;
a memory for storing a computer program;
a processor for implementing the method steps of any of claims 1-7 when executing a program stored on a memory.
16. A computer-readable storage medium, characterized in that a computer program is stored in the computer-readable storage medium, which computer program, when being executed by a processor, carries out the method steps of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810326692.7A CN110378169B (en) | 2018-04-12 | 2018-04-12 | Gesture interval detection method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810326692.7A CN110378169B (en) | 2018-04-12 | 2018-04-12 | Gesture interval detection method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110378169A CN110378169A (en) | 2019-10-25 |
| CN110378169B true CN110378169B (en) | 2021-06-18 |
Family
ID=68243226
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810326692.7A Active CN110378169B (en) | 2018-04-12 | 2018-04-12 | Gesture interval detection method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110378169B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101694692A (en) * | 2009-10-22 | 2010-04-14 | 浙江大学 | Gesture identification method based on acceleration transducer |
| CN101976330A (en) * | 2010-09-26 | 2011-02-16 | 中国科学院深圳先进技术研究院 | Gesture recognition method and system |
| CN103984416A (en) * | 2014-06-10 | 2014-08-13 | 北京邮电大学 | Gesture recognition method based on acceleration sensor |
| US8970479B1 (en) * | 2012-07-31 | 2015-03-03 | Rawles Llc | Hand gesture detection |
| CN105447506A (en) * | 2015-11-05 | 2016-03-30 | 广东省自动化研究所 | Gesture recognition method based on interval distribution probability characteristics |
| CN106293057A (en) * | 2016-07-20 | 2017-01-04 | 西安中科比奇创新科技有限责任公司 | Gesture identification method based on BP neutral net |
| CN107516099A (en) * | 2017-08-21 | 2017-12-26 | 腾讯科技（深圳）有限公司 | A kind of method, apparatus and computer-readable recording medium for marking picture detection |
| CN107728138A (en) * | 2017-09-15 | 2018-02-23 | 电子科技大学 | A kind of maneuvering target tracking method based on current statistical model |
-
2018
- 2018-04-12 CN CN201810326692.7A patent/CN110378169B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101694692A (en) * | 2009-10-22 | 2010-04-14 | 浙江大学 | Gesture identification method based on acceleration transducer |
| CN101976330A (en) * | 2010-09-26 | 2011-02-16 | 中国科学院深圳先进技术研究院 | Gesture recognition method and system |
| US8970479B1 (en) * | 2012-07-31 | 2015-03-03 | Rawles Llc | Hand gesture detection |
| CN103984416A (en) * | 2014-06-10 | 2014-08-13 | 北京邮电大学 | Gesture recognition method based on acceleration sensor |
| CN105447506A (en) * | 2015-11-05 | 2016-03-30 | 广东省自动化研究所 | Gesture recognition method based on interval distribution probability characteristics |
| CN106293057A (en) * | 2016-07-20 | 2017-01-04 | 西安中科比奇创新科技有限责任公司 | Gesture identification method based on BP neutral net |
| CN107516099A (en) * | 2017-08-21 | 2017-12-26 | 腾讯科技（深圳）有限公司 | A kind of method, apparatus and computer-readable recording medium for marking picture detection |
| CN107728138A (en) * | 2017-09-15 | 2018-02-23 | 电子科技大学 | A kind of maneuvering target tracking method based on current statistical model |
Non-Patent Citations (1)
| Title |
|---|
| 手势识别身份认证的连续隐马尔可夫模型;李富 等;《小型微型计算机***》;20180331;第474-477页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110378169A (en) | 2019-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Corner detection based on gradient correlation matrices of planar curves | |
| CN105320937B (en) | Traffic police's gesture identification method based on Kinect | |
| CN108647597B (en) | Wrist identification method, gesture identification method and device and electronic equipment | |
| US20140074431A1 (en) | Wrist Pedometer Step Detection | |
| CN112336342B (en) | Hand key point detection method and device and terminal equipment | |
| CN104049760B (en) | The acquisition methods and system of a kind of man-machine interaction order | |
| CN108846851B (en) | Moving target tracking method and terminal equipment | |
| CN111695429B (en) | Video image target association method and device and terminal equipment | |
| CN104969261B (en) | For detecting the method and system of mobile object | |
| CN111597975B (en) | Personnel action detection method and device and electronic equipment | |
| CN111536989B (en) | Adaptive time window step counting method based on peak detection | |
| EP2967446A2 (en) | Determining activities from sensed motion signals | |
| CN111738225B (en) | Crowd gathering detection method, device, equipment and storage medium | |
| CN108154099B (en) | Figure identification method and device and electronic equipment | |
| CN110688929A (en) | Human skeleton joint point positioning method and device | |
| CN112085701A (en) | Face ambiguity detection method and device, terminal equipment and storage medium | |
| Xi et al. | Device-free human activity recognition using CSI | |
| Liu et al. | 3D Human motion tracking by exemplar-based conditional particle filter | |
| CN109766798A (en) | Gesture data processing method, server and awareness apparatus based on experience small echo | |
| CN116129350A (en) | Intelligent monitoring method, device, equipment and medium for safety operation of data center | |
| CN110378169B (en) | Gesture interval detection method and device | |
| Muhammad et al. | Hand gesture user interface for smart devices based on mems sensors | |
| CN113064118A (en) | Sound source positioning method and device | |
| Iyer et al. | Generalized hand gesture recognition for wearable devices in IoT: Application and implementation challenges | |
| CN112237426A (en) | Method and device for detecting steps, detection equipment and computer 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 |