Methods and systems for identifying user action

Abstract

The embodiment of the present disclosure provides a method and a system for identifying a user action. The method and system may obtain user action data collected from a plurality of measurement positions on a user, the user action data corresponding to an unknown user action, identify that the user action includes a target action when obtaining the user action data based on at least one set of target reference action data, the at least one set of target reference action data corresponding to the target action, and send information related to the target action to the user.

1 . A method for identifying a user action, comprising:
 obtaining user action data collected from a plurality of measurement positions on a user, the user action data corresponding to an unknown user action;   identifying, based on at least one set of target reference action data, that the user action includes a target action when obtaining the user action data, the at least one set of target reference action data corresponding to the target action; and   sending information related to the target action to the user.   
     
     
         2 . The method of  claim 1 , wherein the identifying that the user action includes a target action comprises:
 obtaining a plurality of sets of candidate reference action data, wherein each set of candidate reference action data corresponds to at least one reference action;   performing a two-level screening operation on the plurality of sets of candidate reference action data based on the user action data, the two-level screening operation including a combination of a difference degree-based screening operation and a probability-based screening operation; and   determining that the user action includes the target action based on a result of the two-level screening operation.   
     
     
         3 . The method of  claim 1 , wherein the identifying that the user action includes a target action comprises:
 obtaining a plurality of sets of reference action data, wherein each set of reference action data corresponds to at least one reference action;   selecting each set of reference action data in turn from the plurality of sets of reference action data as candidate reference action data;   determining at least one difference degree by comparing at least one segment of action identification sub-data of the candidate reference action data with the corresponding user action sub-data segment by segment; and   determining a comprehensive difference degree by weighting and summing the at least one difference degree.   
     
     
         4 . The method of  claim 3 , wherein,
 each set of reference action data includes M pieces of reference action sub-data, each piece of the reference action sub-data includes at least one segment of action identification sub-data, and M is an integer greater than 1;   action identification sub-data of the M pieces of reference action sub-data form integral action identification data, and each segment of action identification sub-data corresponds to at least a portion of the reference action on at least one measurement position of the plurality of measurement positions.   
     
     
         5 . The method of  claim 3 , wherein the determining at least one difference degree by comparing at least one segment of action identification sub-data of the candidate reference action data with the corresponding user action sub-data segment by segment comprises:
 selecting a sliding window with a preset length on each piece of the action identification sub-data, the sliding window including a data segment of the user action data collected in a preset time interval; and   for the sliding window at a current moment, determining the difference degree between the data segment and the corresponding action identification sub-data.   
     
     
         6 . The method of  claim 5 , wherein the identifying that the user action includes the target action further comprises:
 determining that a value of the comprehensive difference degree is greater than a first preset value; and   sliding the sliding window to a next data segment with a preset step size, and repeating the comparison.   
     
     
         7 . The method of  claim 6 , wherein a data collection time length corresponding to the data segment in the sliding window is negatively correlated with a user action speed. 
     
     
         8 . The method of  claim 7 , wherein the preset step size satisfies one or more following conditions:
 the preset step size is positively correlated with a magnitude of a value of the comprehensive difference degree at a previous moment; and   the preset step size is positively correlated with a variation trend of the value of the comprehensive difference degree.   
     
     
         9 . The method of  claim 5 , wherein the data segment comprises a plurality of user action data points; and
 the determining at least one difference degree by comparing at least one segment of action identification sub-data of the candidate reference action data with the corresponding user action sub-data segment by segment comprises:
 selecting a target comparison data interval from the action identification sub-data, wherein the target comparison data interval includes a plurality of identification data points, 
 adjusting the data segment according to a plurality of scales to obtain a plurality of adjusted data segments, 
 determining a difference degree between the action identification sub-data and each adjusted data segment of the plurality of adjusted data segments respectively, and 
 determining a minimum difference degree among the difference degrees between the action identification sub-data and the plurality of adjusted data segments. 
   
     
     
         10 . The method of  claim 5 , wherein the determining at least one difference degree by comparing at least one segment of action identification sub-data of the candidate reference action data with the corresponding user action sub-data segment by segment comprises:
 determining a distance matrix [D ij ], wherein D ij  denotes a distance between an i-th data point of a target comparison data interval and a j-th data point of the data segment;   determining a shortest distance path of the distance matrix, wherein the shortest distance path satisfies:   a start point of the shortest distance path being in the first line of the [D ij ],   two adjacent points on the shortest distance path being adjacent in the distance matrix,   a next point on the shortest distance path being to the right, below or right below a previous point,   an end point of the shortest distance path being in a last line of the [D ij ], and   the shortest distance path having a smallest regularization cost, wherein the regularization cost is determined by distances of points on the corresponding shortest distance path of the distance matrix; and   the difference degree being related to the regularization cost.   
     
     
         11 . The method of  claim 10 , wherein if the first data point of the data segment is determined to be a data point where the user action starts, the start point of the shortest distance path is a distance D 11  between the first point of the data segment and the first point of the target comparison data interval. 
     
     
         12 . The method of  claim 10 , wherein if the last data point of the data segment is determined to be the data point where the user action ends, the end point of the shortest distance path is a distance D mn  between the last point of the data segment and the last point of the target comparison data interval. 
     
     
         13 . The method of  claim 3 , wherein the identifying that the user action includes the target action further comprises:
 selecting N pieces of second-level candidate reference action data from the plurality of sets of reference action data, a value of the comprehensive difference degree of the second-level candidate reference action data being less than a first preset value, and N being an integer greater than 1;   calculating N distances between the user action data and the N pieces of second-level candidate reference action data respectively;   calculating N probability values based on the N distances respectively;   selecting the second-level candidate reference action data whose probability value is greater than a second preset value as the target reference action data; and   determining a reference action corresponding to the target reference action data as the target action.   
     
     
         14 . A system for identifying a user action, comprising:
 at least one storage medium, the at least one storage medium storing at least one instruction set for obtaining user action data during the user' motion; and   at least one processor in communication with the at least one storage medium, wherein when the system is running, the at least one processor reads the at least one instruction set and executes the method including:
 obtaining user action data collected from a plurality of measurement positions on a user, the user action data corresponding to an unknown user action; 
 identifying, based on at least one set of target reference action data, that the user action includes a target action when obtaining the user action data, the at least one set of target reference action data corresponding to the target action; and 
 sending information related to the target action to the user. 
   
     
     
         15 . The system of  claim 14 , wherein the identifying that the user action includes a target action comprises:
 obtaining a plurality of sets of candidate reference action data, wherein each set of candidate reference action data corresponds to at least one reference action;   performing a two-level screening operation on the plurality of sets of candidate reference action data based on the user action data, the two-level screening operation including a combination of a difference degree-based screening operation and a probability-based screening operation; and   determining that the user action includes the target action based on a result of the two-level screening operation.   
     
     
         16 . The system of  claim 14 , wherein the identifying that the user action includes a target action comprises:
 obtaining a plurality of sets of reference action data, wherein each set of reference action data corresponds to at least one reference action;   selecting each set of reference action data in turn from the plurality of sets of reference action data as candidate reference action data;   determining at least one difference degree by comparing at least one segment of action identification sub-data of the candidate reference action data with the corresponding user action sub-data segment by segment; and   determining a comprehensive difference degree by weighting and summing the at least one difference degree.   
     
     
         17 . The system of  claim 16 , wherein,
 each set of reference action data includes M pieces of reference action sub-data, each piece of the reference action sub-data includes at least one segment of action identification sub-data, and M is an integer greater than 1;   action identification sub-data of the M pieces of reference action sub-data form integral action identification data, and each segment of action identification sub-data corresponds to at least a portion of the reference action on at least one measurement position of the plurality of measurement positions.   
     
     
         18 . The system of  claim 16 , wherein the determining at least one difference degree by comparing at least one segment of action identification sub-data of the candidate reference action data with the corresponding user action sub-data segment by segment comprises:
 selecting a sliding window with a preset length on each piece of the action identification sub-data, the sliding window including a data segment of the user action data collected in a preset time interval; and   for the sliding window at a current moment, determining the difference degree between the data segment and the corresponding action identification sub-data.   
     
     
         19 . The system of  claim 16 , wherein the identifying that the user action includes the target action further comprises:
 selecting N pieces of second-level candidate reference action data from the plurality of sets of reference action data, a value of the comprehensive difference degree of the second-level candidate reference action data being less than a first preset value, and N being an integer greater than 1;   calculating N distances between the user action data and the N pieces of second-level candidate reference action data respectively;   calculating N probability values based on the N distances respectively;   selecting the second-level candidate reference action data whose probability value is greater than a second preset value as the target reference action data; and   determining a reference action corresponding to the target reference action data as the target action.   
     
     
         20 . A non-transitory computer readable medium, comprising at least one set of instructions for identifying a user action, wherein when executed by at least one processor of a computing device, the at least one set of instructions direct the at least one processor to perform operations including:
 obtaining user action data collected from a plurality of measurement positions on a user, the user action data corresponding to an unknown user action;   identifying, based on at least one set of target reference action data, that the user action includes a target action when obtaining the user action data, the at least one set of target reference action data corresponding to the target action; and   sending information related to the target action to the user.