CN104851301B - Vehicle parameter identification method based on deceleration strip sound analysis - Google Patents

Abstract

The invention relates to a vehicle parameter identification method based on deceleration strip sound analysis. According to the invention, acoustic waveform is acquired by a senor in real time when a vehicle passes two deceleration strips, and the speed of the vehicle is acquired by using the wave crest time and the distance between the two deceleration strips; the wheelbase between two wheels of the vehicle is further acquired on the basis of acquiring the speed of the vehicle, and the acquired wheelbase is compared with basic vehicle type wheelbase information so as to deduce the type of the detected vehicle, thereby finishing vehicle classification. The vehicle parameter identification method only needs one sensor, thereby overcoming a defect of difficult synchronization in a multi-sensor detection environment. The method provided by the invention acquires the acoustic waveform of the deceleration strip easily, mainly completes vehicle speed measurement and vehicle type identification within a time domain, and is simpler and more accurate than other existing methods.

Description

A kind of vehicle parameter recognition methodss based on deceleration strip phonetic analysiss

Technical field

The invention belongs to waveform analyses field, detect speed, knowledge particularly with regard to by analyzing deceleration strip sound waveform The method of the vehicle parameters such as other type of vehicle.

Background technology

Automobile automatic recognition and the important component part that identification is intelligent traffic, in the past using ground waveform analysis, car Grow the method that voice signal etc. carries out vehicle identification, majority is to complete wave character point in multisensor and transform domain Extract；Therefore have the disadvantage in that

1st, under FUSION WITH MULTISENSOR DETECTION environment, the synchronization completing between sensor is relatively difficult；

2nd, more the extraction to waveform, adopted level threshold method in the past, took more；

3rd, the extraction comparison completing characteristic point in transform domain is loaded down with trivial details；

4th, the drifting problem of multisensor is more serious.

Disadvantage mentioned above is easily caused the inaccurate of result.

Content of the invention

The problems referred to above existing for prior art, it is an object of the invention to provide a kind of simple, compared with high-accuracy, reliability The good vehicle identification method based on deceleration strip acoustic waveform analysis of property.

For achieving the above object, the present invention adopts the following technical scheme that a kind of ginseng of the vehicle based on deceleration strip phonetic analysiss Number recognition methodss, comprise the steps:

S1: audio sensor is installed on the road between two deceleration strips；

S2: detect that Current vehicle crosses the road between this two deceleration strips and two deceleration strips by described audio sensor The real-time acoustic waveform being formed during road, records Current vehicle simultaneously and crosses the road between this two deceleration strips and two deceleration strips Time；

S3: the real-time sound wave that obtains of detection is carried out with following denoising:

S3a: described real-time waveform is filtered and smoothing processing；

S3b: judge whether the crest value of the real-time waveform after s3a is processed is more than threshold value a, threshold value a is empirical value, such as The crest value of fruit real-time waveform is more than threshold value a, then this real-time waveform is doubtful waveform, and executes next step；Otherwise return to step s2；

S3c: calculate the rising edge speed of doubtful waveform, be more than threshold value b if up along speed, threshold value b is empirical value, then Determine this doubtful waveform be described vehicle cross between this two deceleration strips and two deceleration strips road when the sound wave ripple that formed Shape, otherwise return to step s2；

S4: the acoustic waveform determining through step s3c has two crest groups, i-th crest pair in first crest group The time point answered is designated asIn second crest group, the corresponding time point of j-th crest is designated asIt is calculated according to formula (1) Speed:

$v=\frac{l}{t_i^2-t_j^1},i=j=1,2,\mathrm{..},d---\left(1\right);$

Wherein v represents speed between two deceleration strips for the vehicle, and l represents the distance between two deceleration strips, and d represents car Total row of wheel.

As optimization, determine the classification of vehicle, step is as follows:

1) wheelbase according to formula (2) the adjacent two row's wheels of calculating:

$s_k=v\·(t_i-t_{i-1})=\frac{l}{t_i^2-t_j^1}\·(t_i-t_{i-1}),i=j---\left(2\right);$

Wherein s_kRepresent the wheelbase of adjacent two row's wheels；

2) the first-to-last of axle dimension s according to formula (3) calculating vehicle:

$s=\underset{k=1}{\overset{d-1}{\σ}}{s}_k---\left(3\right);$

3) the first-to-last of axle dimension s according to vehicle, contrasts vehicle classification wheelbase information, just can get the classification of Current vehicle.

As optimization, described audio sensor is in the centre position of two isolation strip.

With respect to prior art, the present invention has the advantage that

1st, the present invention only needs a sensor, overcomes under FUSION WITH MULTISENSOR DETECTION environment, synchronous difficult shortcoming；

2nd, the method is readily available the live sound ripple that vehicle crosses the road between two deceleration strips and two deceleration strips Shape；

3rd, the mensure of vehicle speed and the identification of vehicle are completed in time domain, more simpler, accurate than other existing methods.

Brief description

Fig. 1 installs illustraton of model for this method sensor, and the dotted arrow of in figure represents garage direction.

Fig. 2 is the overview flow chart of the inventive method.

Fig. 3 is the flow chart to real-time waveform denoising in step s3.

Fig. 4 is the sound waveform figure after processing through step s3.

In figure, 1- first isolation strip, 2- second isolation strip, 3- audio sensor.

Specific embodiment

In describing the invention it is to be understood that term " first ", " second " are only used for describing purpose, and can not It is interpreted as indicating or imply relative importance or the implicit quantity indicating indicated technical characteristic.Thus, define " the One ", the feature of " second " can be expressed or implicitly include one or more this feature.In describing the invention, " multiple " are meant that two or more, unless otherwise expressly limited specifically.

Below the present invention is described in further detail.

Referring to Fig. 1 to Fig. 4, a kind of vehicle parameter recognition methodss based on deceleration strip phonetic analysiss, comprise the steps:

S1: two deceleration strips are selected on highway, the highway between two deceleration strips is then detection zone, this detection zone Including two deceleration strips itself；Audio sensor is installed on the road between two deceleration strips, installs in detection zone Audio sensor；

S2: detect that Current vehicle crosses the road between this two deceleration strips and two deceleration strips by described audio sensor The real-time acoustic waveform being formed during road, records Current vehicle simultaneously and crosses the road between this two deceleration strips and two deceleration strips Time；

S3: the real-time sound wave that obtains of detection is carried out with following denoising:

S3a: described real-time waveform is filtered and smoothing processing, described filtering and smooth be processed as prior art；

S3b: judge whether the crest value of the real-time waveform after s3a is processed is more than threshold value a, threshold value a is empirical value, such as The crest value of fruit real-time waveform is more than threshold value a, then this real-time waveform is doubtful waveform, and executes next step；Otherwise return to step S2, if that is, the crest value of real-time waveform is less than or equal to threshold value a, return to step s2 detects again；

S3c: calculate the rising edge speed of doubtful waveform, be more than threshold value b if up along speed, threshold value b is empirical value, then Determine this doubtful waveform be described vehicle cross between this two deceleration strips and two deceleration strips road when the sound wave ripple that formed Shape, otherwise return to step s2, it is less than or equal to threshold value a if up along speed more than threshold value b, then return to step s2 is examined again Survey；

The computational methods of doubtful waveform rising edge speed are as follows:

1) each doubtful waveform all at least has two crest groups (wheel crosses and often crosses a deceleration strip and will produce one Individual crest group, if vehicle has n row's wheel, then then have n crest in each crest group, the doubtful waveform due to collection may There is noise jamming it is thus possible to have multiple crest groups).

2) determine adjacent with first crest trough in first or second crest group, choose this trough to first ripple Ascending curve section between peak (first crest in first or second crest group), then in this ascending curve section arbitrarily Choose one section as calculating section, using formula (a) calculating rising edge speed:

The vertical coordinate wherein calculating two end points of section is amplitude f, and the abscissa calculating two end points of section is time t.

When being embodied as, it is preferably selected ascending curve Duan Zhongcong trough and is in ascending curve section 10%- to primary peak 90% (rise time, can response waveform change rapidity) part as calculating section, the rising edge speed of determination is more accurate.

Or determine adjacent with last crest trough in first or second crest group, choose last crest Decline curve section to this trough (last crest in first or second crest group), then in this decline curve One section is arbitrarily chosen as calculating section, using formula (a) rising edge speed in section.

When being embodied as, it is preferably selected decline curve Duan Zhongcong trough and is in decline curve section 25%- to last crest 75% part is more accurate as calculating section, the rising edge speed of determination.The acoustic waveform tool that s4: note determines through step s3c There are two crest groups, wherein first crest group is that the wheel of Current vehicle produces when first isolation strip, first wave First crest in peak group be the first row wheel (i.e. vehicle front-wheel) of Current vehicle by first isolation strip when produce, Secondary peak in primary peak group be the second row wheel of Current vehicle by first isolation strip when produce, class successively Push away, send out in primary peak group i-th crest be i-th row's wheel of Current vehicle by first isolation strip when produce；Second Individual crest group is that the wheel of Current vehicle produces when second isolation strip, and first crest in secondary peak group is to work as Produce when the first row wheel (i.e. vehicle front-wheel) of vehicle in front is by second isolation strip, the secondary peak in secondary peak group Be the second row wheel of Current vehicle by second isolation strip when produce, the like, in secondary peak group send out j-th ripple Peak be Current vehicle jth row wheel by second isolation strip when produce；

In first crest group, the corresponding time point of i-th crest is designated asIn second crest group, j-th crest corresponds to Time point be designated asSpeed is calculated according to formula (1):

$v=\frac{l}{t_i^2-t_j^1},i=j=1,2,\mathrm{..},d---\left(1\right);$

Wherein v represents speed between two deceleration strips for the vehicle, and l represents the distance between two deceleration strips, and d represents car Total row of wheel.

As optimization, the inventive method can also further determine that the classification of vehicle, and step is as follows:

1) wheelbase according to formula (2) the adjacent two row's wheels of calculating:

$s_k=v\·(t_i-t_{i-1})=\frac{l}{t_i^2-t_j^1}\·(t_i-t_{i-1}),i=j---\left(2\right);$

Wherein s_kRepresent the wheelbase of adjacent two row's wheels；

2) the first-to-last of axle dimension s according to formula (3) calculating vehicle:

$s=\underset{k=1}{\overset{d-1}{\σ}}{s}_k---\left(3\right);$

3) the first-to-last of axle dimension s according to vehicle, (vehicle classification wheelbase information is existing disclosure to contrast vehicle classification wheelbase information Data), just can get the classification of Current vehicle.

As optimization, audio sensor is in the centre position of two isolation strip,.

In conjunction with Fig. 1, if sound transmission is t in the time of l/2₀, then speed is calculated according to formula (4):

$v=\frac{l}{(t_j^2-t_0)-(t_i^1-t_0)}=\frac{l}{t_j^2-t_i^1},i=j---\left(4\right);$

Formula (4) is identical with formula (1), but this position setting of sensor can overcome the time delay that single-sensor causes to ask Topic, improves the accuracy of detection.

Finally illustrate, above example only in order to technical scheme to be described and unrestricted, although with reference to relatively Good embodiment has been described in detail to the present invention, it will be understood by those within the art that, can be to the skill of the present invention Art scheme is modified or equivalent, the objective without deviating from technical solution of the present invention and scope, and it all should be covered at this In the middle of the right of invention.

Claims (3)

1. a kind of vehicle parameter recognition methodss based on deceleration strip phonetic analysiss are it is characterised in that comprise the steps:

S1: audio sensor is installed on the road between two deceleration strips；

S2: when detecting that Current vehicle crosses the road between this two deceleration strips and two deceleration strips by described audio sensor The real-time acoustic waveform being formed, when record Current vehicle crosses the road between this two deceleration strips and two deceleration strips simultaneously Between；

S3: the real-time sound wave that obtains of detection is carried out with following denoising:

S3a: described real-time waveform is filtered and smoothing processing；

S3b: judge whether the crest value of the real-time waveform after s3a is processed is more than threshold value a, threshold value a is empirical value, such as fruit When waveform crest value be more than threshold value a, then this real-time waveform is doubtful waveform, and executes next step；Otherwise return to step s2；

S3c: calculate doubtful waveform rising edge speed, if up along speed be more than threshold value b, threshold value b be empirical value it is determined that This doubtful waveform be described vehicle cross between this two deceleration strips and two deceleration strips road when the acoustic waveform that formed, no Then return to step s2；

S4: the acoustic waveform determining through step s3c has two crest groups, and in first crest group, i-th crest is corresponding Time point is designated asIn second crest group, the corresponding time point of j-th crest is designated asCar is calculated according to formula (1) Speed:

$v=\frac{l}{t_i^2-t_j^1},i=j=1,2,\mathrm{..},d---\left(1\right);$

Wherein v represents speed between two deceleration strips for the vehicle, and l represents the distance between two deceleration strips, and d represents wheel Total row.

2. the vehicle parameter recognition methodss based on deceleration strip phonetic analysiss as claimed in claim 1 are it is characterised in that determine car Classification, step is as follows:

1) wheelbase according to formula (2) the adjacent two row's wheels of calculating:

$s_k=v\·(t_i-k_{i-1})=\frac{l}{t_i^2-t_j^1}\·(t_i-t_{i-1}),i=j---\left(2\right);$

Wherein s_kRepresent the wheelbase of adjacent two row's wheels；

2) the first-to-last of axle dimension s according to formula (3) calculating vehicle:

$s=\underset{k=1}{\overset{d-1}{\σ}}{s}_k---\left(3\right);$

3) the first-to-last of axle dimension s according to vehicle, contrasts vehicle classification wheelbase information, just can get the classification of Current vehicle.

3. the vehicle parameter recognition methodss based on deceleration strip phonetic analysiss as claimed in claim 1 or 2 are it is characterised in that institute State the centre position that audio sensor is in two isolation strip.