CN103196460B - A kind of method and apparatus estimating step-length - Google Patents

Abstract

The invention discloses a kind of method and apparatus estimating step-length, belong to the communications field.Described method comprises: the first acceleration of the time point that the travel time made a move according to pedestrian's row and described pedestrian comprise at described travel time, obtain fisrt feature amount, described fisrt feature amount comprise the first cadence that described pedestrian's row makes a move, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that described pedestrian's row makes a move; Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively comprises and described fisrt feature amount; Select the second feature amount that the Euclidean distance between described fisrt feature amount is minimum, step-length corresponding for the second feature amount of described selection is defined as the step-length that described pedestrian's row makes a move.Described device comprises: the first acquisition module, the first computing module and determination module.The present invention can reduce the <u> complexity </u> of calculating, and reduces the cost of hardware.

Description

A kind of method and apparatus estimating step-length

Technical field

The present invention relates to the communications field, particularly a kind of method and apparatus estimating step-length.

Background technology

In navigational system, pedestrian's reckoning location technology based on inertial sensor is the developing direction of inertial sensor location navigation, and affect the step-length that one of factor of pedestrian's reckoning precision is pedestrian, in order to improve the precision of pedestrian's reckoning, need the step-length estimating pedestrian.

At present, what provide a kind ofly estimates that the method for step-length is specially: jpeg decompression contract drawing picture is divided into non-overlapping copies and the multiple image block of continuous print, carries out two-dimension discrete cosine transform to each image block, obtains discrete cosine transform coefficient; The step-length that pedestrian's row makes a move is estimated according to discrete cosine transform coefficient.

Realizing in process of the present invention, inventor finds that prior art at least exists following problem:

Two-dimension discrete cosine transform is carried out to each image block and obtains discrete cosine transform coefficient, and estimate that the computation complexity of the step-length that pedestrian's row makes a move is higher according to discrete cosine transform coefficient, and the requirement of prior art to hardware platform is higher, improves hardware cost.

Summary of the invention

In order to solve the problem of prior art, embodiments provide a kind of method and apparatus estimating step-length.Described technical scheme is as follows:

On the one hand, provide a kind of method estimating step-length, described method comprises:

First acceleration of the time point that the travel time made a move according to pedestrian's row and described pedestrian comprise at described travel time, obtain fisrt feature amount, described fisrt feature amount comprise the first cadence that described pedestrian's row makes a move, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that described pedestrian's row makes a move;

Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively comprises and described fisrt feature amount;

Select the second feature amount that the Euclidean distance between described fisrt feature amount is minimum, step-length corresponding for the second feature amount of described selection is defined as the step-length that described pedestrian's row makes a move.

Wherein, the first acceleration of the time point that the described travel time that makes a move according to pedestrian's row and described pedestrian comprise at described travel time, obtains fisrt feature amount, comprising:

According to the travel time that pedestrian's row makes a move, calculate the first cadence that described pedestrian's row makes a move;

First acceleration of the time point comprised at described travel time according to described pedestrian, calculates the first acceleration variance that described pedestrian's row makes a move;

The first acceleration maximal value and the first acceleration minimum value is selected first acceleration of the time point comprised at described travel time from described pedestrian;

By described first cadence, described first acceleration variance, described first acceleration maximal value and described first acceleration minimum value composition fisrt feature amount.

Wherein, the first acceleration of the described time point comprised at described travel time according to described pedestrian, calculates the first acceleration variance that described pedestrian's row makes a move, comprising:

The first acceleration is inputted successively to receive window first the first acceleration that the first step, the first acceleration of time point comprised at described travel time from described pedestrian comprise, until stop input when inputting predetermined number the first acceleration in described receive window, described receive window allows at most to input described predetermined number the first acceleration and described predetermined number is less than the number of the first acceleration of the time point that described pedestrian comprises at described travel time, and the initial value arranging input number is described predetermined number;

Second step, the first acceleration comprised according to described receive window, calculate variance;

If the 3rd step described input number is less than the number of the first acceleration of the time point that described pedestrian comprises at described travel time, then move described receive window, to input next first acceleration to described receive window, and the first acceleration inputted at first comprised by described receive window exports, increase described input number, return and perform described second step;

If the 4th step described input number equals the number of the first acceleration of the time point that described pedestrian comprises at described travel time, then the described variance calculated is averaged value computing, the mean value of the variance calculated described in obtaining, is defined as the first acceleration variance that described pedestrian's row makes a move by described mean value.

Wherein, the Euclidean distance between the second feature amount that the described corresponding relation calculating step-length and the second feature amount stored respectively comprises and described fisrt feature amount, comprising:

Scan from the Article 1 record in the corresponding relation of the step-length stored and second feature amount, obtain the second feature amount in a record of described scanning;

According to the first cadence, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that described fisrt feature amount comprises, and the second cadence, the second acceleration variance, the second acceleration maximal value and the second acceleration minimum value that the second feature amount of described acquisition comprises, according to the Euclidean distance described in following formulae discovery between fisrt feature amount and the second feature amount of described acquisition

$l=\sqrt{{(f-f_i)}^2+{(\mathrm{\&sigma;}-{\mathrm{\&sigma;}}_i)}^2+{(m-m_i)}^2+{(n-n_i)}^2}$

Wherein, in described formula, l is the Euclidean distance between described fisrt feature amount and the second feature amount of described acquisition, f is described first cadence, and σ is described first acceleration variance, and m is described first acceleration maximal value, n is described first acceleration minimum value, f _ifor described second cadence, σ _ifor described second acceleration variance, m _ifor described second acceleration maximal value, n _ifor described second acceleration minimum value;

Judge that whether one of described scanning record is the last item record in the corresponding relation of described step-length and second feature amount, if so, then end operation, otherwise, continue to scan next record.

Further, the first acceleration of the time point that the described travel time that makes a move according to pedestrian's row and described pedestrian comprise at described travel time, before obtaining fisrt feature amount, also comprises:

Obtain the travel time that pedestrian row makes a move, and in three-dimensional system of coordinate, obtain the X direction acceleration of the time point that described pedestrian comprises at described travel time, y direction acceleration and plotted acceleration;

The X direction acceleration of the time point comprised at described travel time according to local acceleration of gravity and described pedestrian, y direction acceleration and plotted acceleration, calculate the first acceleration of the time point that described pedestrian comprises at described travel time respectively.

On the other hand, provide a kind of device estimating step-length, described device comprises:

First acquisition module, first acceleration of the time point comprised at described travel time for the travel time that makes a move according to pedestrian's row and described pedestrian, obtain fisrt feature amount, described fisrt feature amount comprise the first cadence that described pedestrian's row makes a move, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that described pedestrian's row makes a move;

First computing module, the Euclidean distance between the second feature amount that the corresponding relation for calculating step-length and the second feature amount stored respectively comprises and the fisrt feature amount that described first acquisition module obtains;

Determination module, for the second feature amount selecting the Euclidean distance between the fisrt feature amount that described first acquisition module obtains minimum, is defined as the step-length that described pedestrian's row makes a move by step-length corresponding for the second feature amount of described selection.

Wherein, described first acquisition module comprises:

First computing unit, for the travel time made a move according to pedestrian's row, calculates the first cadence that described pedestrian's row makes a move;

Second computing unit, for the first acceleration of time point comprised at described travel time according to described pedestrian, calculates the first acceleration variance that described pedestrian's row makes a move;

Selection unit, selects the first acceleration maximal value and the first acceleration minimum value in the first acceleration of time point of comprising at described travel time from described pedestrian;

Component units, for forming fisrt feature amount by described first cadence, described first acceleration variance, described first acceleration maximal value and described first acceleration minimum value.

Wherein, described second computing unit comprises:

Input subelement, the first acceleration is inputted successively to receive window for first the first acceleration that the first acceleration of the time point comprised at described travel time from described pedestrian comprises, until stop input when inputting predetermined number the first acceleration in described receive window, described receive window allows at most to input described predetermined number the first acceleration and described predetermined number is less than the number of the first acceleration of the time point that described pedestrian comprises at described travel time, and the initial value arranging input number is described predetermined number;

Computation subunit, for being input to the first acceleration that receive window comprises according to described input subelement, calculates variance;

Mover unit, if be less than the number of the first acceleration of the time point that described pedestrian comprises at described travel time for described input number, then move described receive window, to input next first acceleration to described receive window, and the first acceleration inputted at first comprised by described receive window exports, increase described input number, return and perform described computation subunit;

Determine subelement, if equal the number of the first acceleration of the time point that described pedestrian comprises at described travel time for described input number, then the described variance calculated is averaged value computing, the mean value of the variance calculated described in obtaining, is defined as the first acceleration variance that described pedestrian's row makes a move by described mean value.

Wherein, described first computing module comprises:

Acquiring unit, for scanning the Article 1 record in the corresponding relation from the step-length stored and second feature amount, obtains the second feature amount in a record of described scanning;

3rd computing unit, for the first cadence, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that comprise according to described fisrt feature amount, and the second cadence, the second acceleration variance, the second acceleration maximal value and the second acceleration minimum value that the second feature amount of described acquisition comprises, according to the Euclidean distance described in following formulae discovery between fisrt feature amount and the second feature amount of described acquisition

$l=\sqrt{{(f-f_i)}^2+{(\mathrm{\&sigma;}-{\mathrm{\&sigma;}}_i)}^2+{(m-m_i)}^2+{(n-n_i)}^2}$

Wherein, in described formula, l is the Euclidean distance between described fisrt feature amount and the second feature amount of described acquisition, f is described first cadence, and σ is described first acceleration variance, and m is described first acceleration maximal value, n is described first acceleration minimum value, f _ifor described second cadence, σ _ifor described second acceleration variance, m _ifor described second acceleration maximal value, n _ifor described second acceleration minimum value;

Judging unit, for judging that whether one of described scanning record is the last item record in the corresponding relation of described step-length and second feature amount, if so, then end operation, otherwise, continue to scan next record.

Further, described device also comprises:

Second acquisition module, for obtaining the travel time that pedestrian's row makes a move, and obtains the X direction acceleration of the time point that described pedestrian comprises at described travel time, y direction acceleration and plotted acceleration in three-dimensional system of coordinate;

Second computing module, for the X direction acceleration of the time point that comprises at described travel time according to local acceleration of gravity and described pedestrian, y direction acceleration and plotted acceleration, calculate the first acceleration of the time point that described pedestrian comprises at described travel time respectively.

In embodiments of the present invention, the first acceleration of the time point that the travel time made a move according to pedestrian's row and pedestrian comprise at this travel time, obtains fisrt feature amount; Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively comprises and fisrt feature amount; Select the second feature amount that the Euclidean distance between fisrt feature amount is minimum, step-length corresponding for the second feature amount of selection is defined as the step-length that pedestrian's row makes a move.Wherein, fisrt feature amount comprises the first cadence that pedestrian's row makes a move, pedestrian's row makes a move the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value, according to the corresponding relation of fisrt feature amount and the step-length stored and second feature amount, calculate the step-length that pedestrian's row makes a move, the complexity of calculating can be reduced, and reduce hardware cost.

Accompanying drawing explanation

In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.

Fig. 1 is a kind of method flow diagram estimating step-length that the embodiment of the present invention one provides;

Fig. 2 is a kind of method flow diagram estimating step-length that the embodiment of the present invention two provides;

Fig. 3 is a kind of apparatus structure schematic diagram estimating step-length that the embodiment of the present invention three provides.

Embodiment

For making the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, embodiment of the present invention is described further in detail.

Embodiment one

Embodiments provide a kind of method estimating step-length, see Fig. 1, the method comprises:

Step 101: the travel time made a move according to pedestrian's row and pedestrian when walking between the first acceleration of time point of comprising, obtain fisrt feature amount, this fisrt feature amount comprise the first cadence that pedestrian's row makes a move, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that pedestrian's row makes a move;

Step 102: the Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively comprises and fisrt feature amount;

Step 103: select the second feature amount that the Euclidean distance between fisrt feature amount is minimum, step-length corresponding for the second feature amount of selection is defined as the step-length that pedestrian's row makes a move.

In embodiments of the present invention, the first acceleration of the time point that the travel time made a move according to pedestrian's row and pedestrian comprise at this travel time, obtains fisrt feature amount; Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively comprises and fisrt feature amount; Select the second feature amount that the Euclidean distance between fisrt feature amount is minimum, step-length corresponding for the second feature amount of selection is defined as the step-length that pedestrian's row makes a move.Wherein, fisrt feature amount comprises the first cadence that pedestrian's row makes a move, pedestrian's row makes a move the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value, according to the corresponding relation of fisrt feature amount and the step-length stored and second feature amount, calculate the step-length that pedestrian's row makes a move, the complexity of calculating can be reduced, and reduce hardware cost.

Embodiment two

Embodiments provide a kind of method estimating step-length, see Fig. 2, the method comprises:

Step 201: obtain the travel time that pedestrian row makes a move, and in three-dimensional system of coordinate, obtain the X direction acceleration of the time point that pedestrian comprises at this travel time, y direction acceleration and plotted acceleration;

Particularly, obtain the travel time that pedestrian's row makes a move, and according to the travel time that this pedestrian's row makes a move, the X direction acceleration of the time point that pedestrian comprises at this travel time is obtained in pedestrian's X direction Acceleration pulse of three-dimensional system of coordinate, according to the travel time that this pedestrian's row makes a move, the y direction acceleration of the time point that pedestrian comprises at this travel time is obtained in pedestrian's y direction Acceleration pulse of three-dimensional system of coordinate, and according to the travel time that this pedestrian's row makes a move, the plotted acceleration of the time point that pedestrian comprises at this travel time is obtained in pedestrian's plotted Acceleration pulse of three-dimensional system of coordinate.

Wherein, the transverse axis in three-dimensional system of coordinate is X-axis, and the longitudinal axis is Y-axis, and vertical pivot is Z axis.

Step 202: the X direction acceleration of the time point comprised at this travel time according to local acceleration of gravity and pedestrian, y direction acceleration and plotted acceleration, calculates the first acceleration of the time point that pedestrian comprises at this travel time respectively;

Particularly, the X direction acceleration of the time point comprised at this travel time according to local acceleration of gravity and pedestrian, y direction acceleration and plotted acceleration, calculate the first acceleration of the time point that pedestrian comprises at this travel time respectively according to following formula (1);

$a\left(t\right)=\sqrt{a{\left(t\right)}_x^2+a{\left(t\right)}_y^2+a{\left(t\right)}_z^2}-g---\left(1\right)$

Wherein, in formula (1), the first acceleration of the time point that a (t) comprises at this travel time for pedestrian, a (t) _xfor the X direction acceleration of the time point that pedestrian comprises at this travel time, a (t) _yfor the y direction acceleration of the time point that pedestrian comprises at this travel time, a (t) _zfor the plotted acceleration of the time point that pedestrian comprises at this travel time, g is local acceleration of gravity.

Step 203: the travel time made a move according to pedestrian's row, calculates the first cadence that pedestrian's row makes a move;

Particularly, inverse is got to the travel time that pedestrian's row makes a move, obtain the first cadence that pedestrian's row makes a move.

Such as, the travel time that pedestrian's row makes a move is 1s, then get inverse to the travel time 1s that this pedestrian's row makes a move, and obtaining the first cadence that this pedestrian's row makes a move is 1Hz.

Step 204: the first acceleration of the time point comprised at this travel time according to pedestrian, calculates pedestrian's row and to make a move the first acceleration variance;

Particularly, this step can be divided into the step of (1)-(4) as follows, comprising:

(1) the first acceleration is inputted successively to receive window first the first acceleration that the first acceleration of the time point, comprised from pedestrian at this travel time comprises, until stop input when inputting predetermined number the first acceleration in this receive window, this receive window allows at most to input predetermined number the first acceleration and predetermined number is less than the number of the first acceleration of the time point that pedestrian comprises at this travel time, and the initial value arranging input number is predetermined number;

Particularly, before receive window being placed on first the first acceleration of the first acceleration of the time point that pedestrian comprises at this travel time, movably receiving window, to input predetermined number the first acceleration to receive window, the initial value arranging input number is predetermined number, this receive window allows at most to receive predetermined number the first acceleration, and predetermined number is less than the number of the first acceleration of the time point that pedestrian comprises at this travel time.

Wherein, number is inputted for recording the number of the first acceleration be input in receive window.

Further, before this step of execution, also create receive window, and the window size arranging receive window is predetermined number; The window size of receive window is predetermined number, represent in receive window and can hold at most predetermined number the first acceleration, when receive window comprises predetermined number the first acceleration, if input first acceleration to receive window again, then the first acceleration inputted at first that receive window comprises will shift out receive window.

Such as, predetermined number is 3, and the number of the first acceleration of the time point that pedestrian comprises at this travel time is 6, and the sequence of the first acceleration of time point that pedestrian comprises at this travel time is 1m/s ², 8m/s ², 7m/s ², 6m/s ², 4m/s ², 2m/s ², receive window is placed on first acceleration 1m/s ²before, movably receiving window, inputs first the first acceleration 1m/s in receive window ², until stop input when inputting 3 the first acceleration in this receive window, arranging the first acceleration input number initial value is 3, and namely this receive window can receive at most 3 the first acceleration.

(2) the first acceleration, according to receive window comprised, calculates variance;

Particularly, first acceleration in first acceleration of the time point comprised from this travel time of pedestrian, the number of the first acceleration comprised when receive window is predetermined number, according to predetermined number the first acceleration that receive window comprises, calculate the current variance being input to the first acceleration of receive window.

Such as, current the first acceleration being input to receive window is 1m/s ², 8m/s ²and 7m/s ², according to the first acceleration 1m/s of receive window ², 8m/s ²and 7m/s ², calculating the current variance being input to the first acceleration of receive window is 9.57.

(3), judge to input the number whether number reaches the first acceleration of the time point that pedestrian comprises at this travel time, if do not reached, then movably receiving window, to input next first acceleration to receive window, and the first acceleration inputted at first comprised by receive window exports, and increases input number, returns and perform step (2), otherwise, perform step (4);

(4), to the variance calculated be averaged value computing, obtains the mean value of the variance calculated, and this mean value is defined as pedestrian's row and makes a move the first acceleration variance.

Such as, the first acceleration 1m/s ², 8m/s ²and 7m/s ²variance be the 9.57, first acceleration 8m/s ², 7m/s ²and 6m/s ²variance be the 0.67, first acceleration 7m/s ², 6m/s ²and 4m/s ²variance be the 1.56, first acceleration 6m/s ², 4m/s ²be 2.67 with the variance of 2m/s2, to be averaged value computing to the variance 9.57,0.67,1.56 and 2.67 calculated, the mean value obtaining this variance is 3.62, this mean value 3.62 is defined as pedestrian's row and makes a move the first acceleration variance.

Step 205: select the first acceleration maximal value and the first acceleration minimum value the first acceleration of the time point comprised at this travel time from pedestrian;

Particularly, the first acceleration of time point pedestrian comprised at this travel time compares, and selects the first acceleration maximal value and the first acceleration minimum value the first acceleration of the time point comprised at this travel time from this pedestrian.

Step 206: the first cadence that pedestrian's row is made a move, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value composition fisrt feature amount;

Such as, first cadence is 0.1, first acceleration variance is the 3.62, first acceleration maximal value be the 8, first acceleration minimum value is 1, be 0.1 by the first cadence, first acceleration variance to be the 3.62, first acceleration maximal value be 8 and first acceleration minimum value be 1 composition fisrt feature amount (0.1,3.62,8,1).

Step 207: the Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively comprises and fisrt feature amount;

Particularly, this step can be divided into the step of (1)-(3) as follows, comprising:

(1), scan from the Article 1 record in the corresponding relation of the step-length stored and second feature amount;

(2), the second feature amount obtained in a record of scanning, and the Euclidean distance between the second feature amount calculating fisrt feature amount and acquisition;

Particularly, according to the first cadence, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that fisrt feature amount comprises, and the second cadence, the second acceleration variance, the second acceleration maximal value and the second acceleration minimum value that the second feature amount obtained comprises, the Euclidean distance between fisrt feature amount and the second feature amount of acquisition is calculated according to following formula (2)

$l=\sqrt{{(f-f_i)}^2+{(\mathrm{\&sigma;}-{\mathrm{\&sigma;}}_i)}^2+{(m-m_i)}^2+{(n-n_i)}^2}---\left(2\right)$

Wherein, in formula (2), l is the Euclidean distance between fisrt feature amount and second feature amount, and f is the first cadence, and σ is the first acceleration variance, and m is the first acceleration maximal value, and n is the first acceleration minimum value, f _ibe the second cadence, σ _ibe the second acceleration variance, m _ibe the second acceleration maximal value, n _iit is the second acceleration minimum value.

(3), judge that whether the record scanned is the last item record in the corresponding relation of step-length and second feature amount, if so, then end operation, otherwise, continue next record of scanning, return execution step (2).

Wherein, a prior selection default value step-length, be directed to the some step-lengths in a default value step-length of selection, pedestrian walks according to this step-length, obtain the step number of pedestrian's walking, and obtain the second acceleration of the time point that the travel time of pedestrian and pedestrian comprise at this travel time, by the walking step number of pedestrian that obtains divided by the travel time obtained, obtain the second cadence that pedestrian is directed to this step-length; According to the second acceleration of the time point that the travel time of the pedestrian obtained comprises at this travel time, calculate the second acceleration variance that pedestrian is directed to this step-length; Obtain acceleration maximal value of each step of pedestrian's walking, the acceleration maximal value obtained is averaged value computing, obtains the second acceleration maximal value; And obtain acceleration minimum value of each step of pedestrian's walking, the acceleration minimum value obtained is averaged value computing, obtains the second acceleration minimum value; Second cadence, the second acceleration variance, the second acceleration maximal value and the second acceleration minimum value composition pedestrian is directed to the second feature amount of this step-length, stores the corresponding relation of this step-length and second feature amount.

Such as, the corresponding relation of the step-length stored and second feature amount, as following table 1, scans from the Article 1 record of table 1, and the second feature amount obtained in the Article 1 record of scanning is (0.2,4,7,1.5)

Table 1

Step-length	Second feature amount
		0.4	（0.2，4，7，1.5）
0.575	（0.23，3.55，10，2）
		0.625	（0.1，3.55，8，1）
0.675	（0.3，3.5，9，2）
		0.8	（0.15，4.2，7，1）

According to the first cadence 0.1, first acceleration variance 3.62, first acceleration maximal value 8 and the first acceleration minimum value 1 that fisrt feature amount comprises, and the second cadence 0.2, second acceleration variance 4, second acceleration maximal value 7 and the second acceleration minimum value 1.5 that the second feature amount obtained comprises, the Euclidean distance calculated between fisrt feature amount and the second feature amount of acquisition according to above-mentioned formula (2) is 0.39.

Step 208: select the second feature amount that the Euclidean distance between fisrt feature amount is minimum, step-length corresponding for the second feature amount of selection is defined as the step-length that pedestrian's row makes a move.

Particularly, select the second feature amount that the Euclidean distance between fisrt feature amount is minimum, and according to the second feature amount selected, from the step-length that the step-length stored is corresponding with the second feature amount obtaining selection the corresponding relation of second feature amount, the step-length of acquisition is defined as the step-length that pedestrian's row makes a move.

Such as, that select and that Euclidean distance between fisrt feature amount is minimum second feature amount is (0.1,3.55,8,1), according to the second feature amount (0.1 selected, 3.55,8,1), the step-length corresponding with the second feature amount obtaining selection the corresponding relation of second feature amount from the step-length stored is 0.625, the step-length 0.625 of acquisition is defined as the step-length that pedestrian's row makes a move.

In embodiments of the present invention, the first acceleration of the time point that the travel time made a move according to pedestrian's row and pedestrian comprise at this travel time, obtains fisrt feature amount; Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively comprises and fisrt feature amount; Select the second feature amount that the Euclidean distance between fisrt feature amount is minimum, step-length corresponding for the second feature amount of selection is defined as the step-length that pedestrian's row makes a move.Wherein, fisrt feature amount comprises the first cadence that pedestrian's row makes a move, pedestrian's row makes a move the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value, according to the corresponding relation of fisrt feature amount and the step-length stored and second feature amount, calculate the step-length that pedestrian's row makes a move, the complexity of calculating can be reduced, and reduce hardware cost.

Embodiment three

See Fig. 3, embodiments provide a kind of device estimating step-length, this device comprises:

First acquisition module 301, for the travel time that makes a move according to pedestrian's row and pedestrian when walking between the first acceleration of time point of comprising, obtain fisrt feature amount, fisrt feature amount comprise the first cadence that pedestrian's row makes a move, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that pedestrian's row makes a move;

First computing module 302, the Euclidean distance between the second feature amount that the corresponding relation for calculating step-length and the second feature amount stored respectively comprises and the fisrt feature amount that the first acquisition module 301 obtains;

Determination module 303, for the second feature amount selecting the Euclidean distance between the fisrt feature amount that the first acquisition module 301 obtains minimum, is defined as the step-length that pedestrian's row makes a move by step-length corresponding for the second feature amount of selection.

Wherein, the first acquisition module 301 comprises:

First computing unit, for the travel time made a move according to pedestrian's row, calculates the first cadence that pedestrian's row makes a move;

Second computing unit, for the first acceleration of time point comprised at this travel time according to pedestrian, calculates the first acceleration variance that pedestrian's row makes a move;

Selection unit, selects the first acceleration maximal value and the first acceleration minimum value in the first acceleration of time point of comprising at this travel time from pedestrian;

Component units, for forming fisrt feature amount by the first cadence, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value.

Further, the second computing unit comprises:

Input subelement, the first acceleration is inputted successively to receive window for first the first acceleration that the first acceleration of the time point comprised at this travel time from pedestrian comprises, until stop input when inputting predetermined number the first acceleration in receive window, receive window allows at most to input predetermined number the first acceleration and predetermined number is less than the number of the first acceleration of the time point that pedestrian comprises at this travel time, and the initial value arranging input number is predetermined number;

Computation subunit, for being input to the first acceleration that receive window comprises according to input subelement, calculates variance;

Mover unit, if be less than the number of the first acceleration of the time point that pedestrian comprises at this travel time for input number, then movably receiving window, to input next first acceleration to receive window, and the first acceleration inputted at first comprised by receive window exports, increase input number, return execution computation subunit;

Determine subelement, if equal the number of the first acceleration of the time point that pedestrian comprises at this travel time for input number, then the variance calculated is averaged value computing, obtain the mean value of the variance calculated, mean value is defined as the first acceleration variance that pedestrian's row makes a move.

Wherein, the first computing module 302 comprises:

Acquiring unit, for scanning the Article 1 record in the corresponding relation from the step-length stored and second feature amount, obtains the second feature amount in a record of scanning;

3rd computing unit, for the first cadence, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that comprise according to fisrt feature amount, and the second cadence, the second acceleration variance, the second acceleration maximal value and the second acceleration minimum value that the second feature amount obtained comprises, according to the Euclidean distance between following formulae discovery fisrt feature amount and the second feature amount of acquisition

$l=\sqrt{{(f-f_i)}^2+{(\mathrm{\&sigma;}-{\mathrm{\&sigma;}}_i)}^2+{(m-m_i)}^2+{(n-n_i)}^2}$

Wherein, in formula, l is the Euclidean distance between the second feature amount of fisrt feature amount and acquisition, and f is the first cadence, and σ is the first acceleration variance, and m is the first acceleration maximal value, and n is the first acceleration minimum value, f _ibe the second cadence, σ _ibe the second acceleration variance, m _ibe the second acceleration maximal value, n _iit is the second acceleration minimum value;

Judging unit, for judging that whether the record scanned is the last item record in the corresponding relation of step-length and second feature amount, if so, then end operation, otherwise, continue to scan next record.

Further, this device also comprises:

Second acquisition module, for obtaining the travel time that pedestrian's row makes a move, and obtain in three-dimensional system of coordinate pedestrian when walking between the X direction acceleration of time point, y direction acceleration and the plotted acceleration that comprise;

Second computing module, for according to local acceleration of gravity and pedestrian when walking between the X direction acceleration of time point, y direction acceleration and the plotted acceleration that comprise, calculate respectively pedestrian when walking between the first acceleration of time point of comprising.

In embodiments of the present invention, the first acceleration of the time point that the travel time made a move according to pedestrian's row and pedestrian comprise at this travel time, obtains fisrt feature amount; Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively comprises and fisrt feature amount; Select the second feature amount that the Euclidean distance between fisrt feature amount is minimum, step-length corresponding for the second feature amount of selection is defined as the step-length that pedestrian's row makes a move.Wherein, fisrt feature amount comprises the first cadence that pedestrian's row makes a move, pedestrian's row makes a move the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value, according to the corresponding relation of fisrt feature amount and the step-length stored and second feature amount, calculate the step-length that pedestrian's row makes a move, the complexity of calculating can be reduced, and reduce hardware cost.

The invention described above embodiment sequence number, just to describing, does not represent the quality of embodiment.

One of ordinary skill in the art will appreciate that all or part of step realizing above-described embodiment can have been come by hardware, the hardware that also can carry out instruction relevant by program completes, described program can be stored in a kind of computer-readable recording medium, the above-mentioned storage medium mentioned can be ROM (read-only memory), disk or CD etc.

The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (10)

1. estimate a method for step-length, it is characterized in that, described method comprises:

First acceleration of the time point that the travel time made a move according to pedestrian's row and described pedestrian comprise at described travel time, obtain fisrt feature amount, described fisrt feature amount comprise the first cadence that described pedestrian's row makes a move, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that described pedestrian's row makes a move;

According to the first cadence, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that described fisrt feature amount comprises, and the second cadence, the second acceleration variance, the second acceleration maximal value and the second acceleration minimum value in the second feature amount that comprises of the corresponding relation of the step-length stored and second feature amount, the Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively according to following formula comprises and described fisrt feature amount;

$l=\sqrt{{(f-f_i)}^2+{(\mathrm{\&sigma;}-{\mathrm{\&sigma;}}_i)}^2+{(m-m_i)}^2+{(n-n_i)}^2}$

Wherein, in described formula, l is the Euclidean distance between the second feature amount that comprises of the corresponding relation of step-length and the second feature amount stored and described fisrt feature amount, f is described first cadence, σ is described first acceleration variance, m is described first acceleration maximal value, and n is described first acceleration minimum value, f _ifor described second cadence, σ _ifor described second acceleration variance, m _ifor described second acceleration maximal value, n _ifor described second acceleration minimum value;

Select the second feature amount that the Euclidean distance between described fisrt feature amount is minimum, step-length corresponding for the second feature amount of described selection is defined as the step-length that described pedestrian's row makes a move.

2. method according to claim 1, is characterized in that, the first acceleration of the time point that the described travel time that makes a move according to pedestrian's row and described pedestrian comprise at described travel time, obtains fisrt feature amount, comprising:

According to the travel time that pedestrian's row makes a move, calculate the first cadence that described pedestrian's row makes a move;

First acceleration of the time point comprised at described travel time according to described pedestrian, calculates the first acceleration variance that described pedestrian's row makes a move;

The first acceleration maximal value and the first acceleration minimum value is selected first acceleration of the time point comprised at described travel time from described pedestrian;

By described first cadence, described first acceleration variance, described first acceleration maximal value and described first acceleration minimum value composition fisrt feature amount.

3. method according to claim 2, is characterized in that, the first acceleration of the described time point comprised at described travel time according to described pedestrian, calculates the first acceleration variance that described pedestrian's row makes a move, comprising:

The first acceleration is inputted successively to receive window first the first acceleration that the first step, the first acceleration of time point comprised at described travel time from described pedestrian comprise, until stop input when inputting predetermined number the first acceleration in described receive window, described receive window allows at most to input described predetermined number the first acceleration and described predetermined number is less than the number of the first acceleration of the time point that described pedestrian comprises at described travel time, and the initial value arranging input number is described predetermined number;

Second step, the first acceleration comprised according to described receive window, calculate variance;

If the 3rd step described input number is less than the number of the first acceleration of the time point that described pedestrian comprises at described travel time, then move described receive window, to input next first acceleration to described receive window, and the first acceleration inputted at first comprised by described receive window exports, increase described input number, return and perform described second step;

If the 4th step described input number equals the number of the first acceleration of the time point that described pedestrian comprises at described travel time, then the described variance calculated is averaged value computing, the mean value of the variance calculated described in obtaining, is defined as the first acceleration variance that described pedestrian's row makes a move by described mean value.

4. method according to claim 1, it is characterized in that, described the first cadence comprised according to described fisrt feature amount, first acceleration variance, first acceleration maximal value and the first acceleration minimum value, and the second cadence in the second feature amount that comprises of the corresponding relation of the step-length stored and second feature amount, second acceleration variance, second acceleration maximal value and the second acceleration minimum value, Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively according to following formula comprises and described fisrt feature amount, comprise:

Scan from the Article 1 record in the corresponding relation of the step-length stored and second feature amount, obtain the second feature amount in a record of described scanning;

According to the first cadence, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that described fisrt feature amount comprises, and the second cadence, the second acceleration variance, the second acceleration maximal value and the second acceleration minimum value that the second feature amount of described acquisition comprises, according to the Euclidean distance described in described formulae discovery between fisrt feature amount and the second feature amount of described acquisition;

Judge that whether one of described scanning record is the last item record in the corresponding relation of described step-length and second feature amount, if so, then end operation, otherwise, continue to scan next record.

5. method according to claim 1, is characterized in that, the first acceleration of the time point that the described travel time that makes a move according to pedestrian's row and described pedestrian comprise at described travel time, before obtaining fisrt feature amount, also comprises:

Obtain the travel time that pedestrian row makes a move, and in three-dimensional system of coordinate, obtain the X direction acceleration of the time point that described pedestrian comprises at described travel time, y direction acceleration and plotted acceleration;

The X direction acceleration of the time point comprised at described travel time according to local acceleration of gravity and described pedestrian, y direction acceleration and plotted acceleration, calculate the first acceleration of the time point that described pedestrian comprises at described travel time respectively.

6. estimate a device for step-length, it is characterized in that, described device comprises:

First acquisition module, first acceleration of the time point comprised at described travel time for the travel time that makes a move according to pedestrian's row and described pedestrian, obtain fisrt feature amount, described fisrt feature amount comprise the first cadence that described pedestrian's row makes a move, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that described pedestrian's row makes a move;

First computing module, for the first cadence, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that comprise according to described fisrt feature amount, and the second cadence, the second acceleration variance, the second acceleration maximal value and the second acceleration minimum value in the second feature amount that comprises of the corresponding relation of the step-length stored and second feature amount, the Euclidean distance between the second feature amount that the corresponding relation calculating step-length and the second feature amount stored respectively according to following formula comprises and the fisrt feature amount that described first acquisition module obtains;

$l=\sqrt{{(f-f_i)}^2+{(\mathrm{\&sigma;}-{\mathrm{\&sigma;}}_i)}^2+{(m-m_i)}^2+{(n-n_i)}^2}$

Wherein, in described formula, l is the Euclidean distance between the second feature amount that comprises of the corresponding relation of step-length and the second feature amount stored and described fisrt feature amount, f is described first cadence, σ is described first acceleration variance, m is described first acceleration maximal value, and n is described first acceleration minimum value, f _ifor described second cadence, σ _ifor described second acceleration variance, m _ifor described second acceleration maximal value, n _ifor described second acceleration minimum value;

Determination module, for the second feature amount selecting the Euclidean distance between the fisrt feature amount that described first acquisition module obtains minimum, is defined as the step-length that described pedestrian's row makes a move by step-length corresponding for the second feature amount of described selection.

7. device according to claim 6, is characterized in that, described first acquisition module comprises:

First computing unit, for the travel time made a move according to pedestrian's row, calculates the first cadence that described pedestrian's row makes a move;

Second computing unit, for the first acceleration of time point comprised at described travel time according to described pedestrian, calculates the first acceleration variance that described pedestrian's row makes a move;

Selection unit, selects the first acceleration maximal value and the first acceleration minimum value in the first acceleration of time point of comprising at described travel time from described pedestrian;

Component units, for forming fisrt feature amount by described first cadence, described first acceleration variance, described first acceleration maximal value and described first acceleration minimum value.

8. device according to claim 7, is characterized in that, described second computing unit comprises:

Input subelement, the first acceleration is inputted successively to receive window for first the first acceleration that the first acceleration of the time point comprised at described travel time from described pedestrian comprises, until stop input when inputting predetermined number the first acceleration in described receive window, described receive window allows at most to input described predetermined number the first acceleration and described predetermined number is less than the number of the first acceleration of the time point that described pedestrian comprises at described travel time, and the initial value arranging input number is described predetermined number;

Computation subunit, for being input to the first acceleration that receive window comprises according to described input subelement, calculates variance;

Mover unit, if be less than the number of the first acceleration of the time point that described pedestrian comprises at described travel time for described input number, then move described receive window, to input next first acceleration to described receive window, and the first acceleration inputted at first comprised by described receive window exports, increase described input number, return and perform described computation subunit;

Determine subelement, if equal the number of the first acceleration of the time point that described pedestrian comprises at described travel time for described input number, then the described variance calculated is averaged value computing, the mean value of the variance calculated described in obtaining, is defined as the first acceleration variance that described pedestrian's row makes a move by described mean value.

9. device according to claim 6, is characterized in that, described first computing module comprises:

Acquiring unit, for scanning the Article 1 record in the corresponding relation from the step-length stored and second feature amount, obtains the second feature amount in a record of described scanning;

3rd computing unit, for the first cadence, the first acceleration variance, the first acceleration maximal value and the first acceleration minimum value that comprise according to described fisrt feature amount, and the second cadence, the second acceleration variance, the second acceleration maximal value and the second acceleration minimum value that the second feature amount of described acquisition comprises, according to the Euclidean distance described in described formulae discovery between fisrt feature amount and the second feature amount of described acquisition;

Judging unit, for judging that whether one of described scanning record is the last item record in the corresponding relation of described step-length and second feature amount, if so, then end operation, otherwise, continue to scan next record.

10. device according to claim 6, is characterized in that, described device also comprises:

Second acquisition module, for obtaining the travel time that pedestrian's row makes a move, and obtains the X direction acceleration of the time point that described pedestrian comprises at described travel time, y direction acceleration and plotted acceleration in three-dimensional system of coordinate;

Second computing module, for the X direction acceleration of the time point that comprises at described travel time according to local acceleration of gravity and described pedestrian, y direction acceleration and plotted acceleration, calculate the first acceleration of the time point that described pedestrian comprises at described travel time respectively.