CN104851301A - Vehicle parameter identification method based on deceleration strip sound analysis - Google Patents

Abstract

The invention relates to a vehicle parameter recognition method based on the sound analysis of speed bumps, which collects the sound wave waveform of the vehicle passing through two speed bumps in real time by a sensor; uses the peak time and the distance between the two speed bumps to obtain the vehicle's Vehicle speed: On the basis of the obtained vehicle speed, the wheelbase between the two wheels of the vehicle is further obtained. After the wheelbase is obtained, the model of the inspected vehicle is inferred by comparison with the wheelbase information of the basic vehicle type, and the vehicle classification is completed. The invention only needs one sensor, which overcomes the disadvantage of difficult synchronization in the multi-sensor detection environment; the method is easier to obtain the sound waveform of the speed bump, and mainly completes the measurement of the vehicle speed and the identification of the vehicle type in the time domain, which is more efficient than other existing methods Simple and accurate.

Description

A Vehicle Parameter Recognition Method Based on Speed Bump Sound Analysis

technical field

The invention belongs to the field of waveform analysis, and in particular relates to a method for detecting vehicle speed, identifying vehicle type and other vehicle parameters by analyzing the sound waveform of the speed bump.

Background technique

Automatic classification and recognition of vehicles is an important part of intelligent transportation. In the past, the methods of using geomagnetic waveform analysis and vehicle-generated voice signals for vehicle recognition mostly completed the extraction of waveform feature points in multi-sensor and transform domains; therefore, there are the following disadvantages :

1. In a multi-sensor detection environment, it is difficult to complete the synchronization between sensors;

2. For the extraction of waveforms, in the past, the multi-level threshold method was often used, which took a lot of time;

3. It is cumbersome to complete the extraction of feature points in the transform domain;

4. The drift problem of multi-sensor is serious.

The above shortcomings can easily lead to inaccurate results.

Contents of the invention

In view of the above-mentioned problems existing in the prior art, the object of the present invention is to provide a simple, high-accuracy, and reliable vehicle identification method based on the sound waveform analysis of speed bumps.

In order to achieve the above object, the present invention adopts the following technical solution: a vehicle parameter recognition method based on speed bump sound analysis, comprising the following steps:

S1: Install an audio sensor on the road between two speed bumps;

S2: Use the audio sensor to detect the real-time sound wave waveform formed when the current vehicle passes the two deceleration belts and the road between the two deceleration belts, and record the current vehicle passing the two deceleration belts and the distance between the two deceleration belts at the same time. road time between

S3: Perform the following denoising processing on the detected real-time sound waves:

S3a: filtering and smoothing the real-time waveform;

S3b: Determine whether the peak value of the real-time waveform processed by S2a is greater than the threshold value A, the threshold value A is an empirical value, if the peak value of the real-time waveform is greater than the threshold value A, then the real-time waveform is a suspected waveform, and perform the next step; otherwise, return to the step S2;

S3c: Calculate the rising edge rate of the suspected waveform. If the rising edge rate is greater than the threshold B, which is an empirical value, then it is determined that the suspected waveform is when the vehicle passes the two speed bumps and the road between the two speed bumps Formed acoustic waveform, otherwise return to step S2;

S4: The sound wave waveform determined in step S3c has two peak groups, and the time point corresponding to the i-th peak in the first peak group is recorded as The time point corresponding to the jth peak in the second peak group is recorded as The vehicle speed is calculated according to the formula (1):

$\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; L</span>}}{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; D.</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$

Among them, v represents the speed of the vehicle between the two speed reduction belts, L represents the distance between the two speed reduction belts, and D represents the total number of rows of wheels.

As an optimization, to determine the classification of the vehicle, the steps are as follows:

1) Calculate the wheelbase of two adjacent rows of wheels according to formula (2):

${\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; L</span>}}{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; \&CenterDot;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$

Where s _k represents the wheelbase of two adjacent rows of wheels;

2) Calculate the total wheelbase s of the vehicle according to formula (3):

$\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; D.</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$

3) According to the total wheelbase s of the vehicle, the classification of the current vehicle can be obtained by comparing the vehicle classification wheelbase information.

As an optimization, the audio sensor is located in the middle of the two isolation zones.

Compared with the prior art, the present invention has the following advantages:

1. The present invention only needs one sensor, which overcomes the disadvantage of difficult synchronization in the multi-sensor detection environment;

2. This method is easy to obtain the real-time sound waveform of the vehicle passing through two deceleration belts and the road between the two deceleration belts;

3. The measurement of the vehicle speed and the identification of the vehicle type are completed in the time domain, which is simpler and more accurate than other existing methods.

Description of drawings

Fig. 1 is the installation model diagram of the sensor of this method, and the dotted arrow in the figure indicates the driving direction.

Fig. 2 is the overall flowchart of the method of the present invention.

FIG. 3 is a flowchart of the real-time waveform denoising processing in step S3.

Fig. 4 is a waveform diagram of the sound processed in step S3.

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

Detailed ways

In the description of the present invention, it should be understood that the terms "first" and "second" are used for description purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly indicating the quantity of indicated technical features. Thus, a feature defined as "first" and "second" may explicitly or implicitly include one or more of these features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

The present invention will be described in further detail below.

Referring to Figures 1 to 4, a vehicle parameter recognition method based on sound analysis of speed bumps includes the following steps:

S1: Select two deceleration belts on the highway, and the road between the two deceleration belts is the detection area, which includes the two deceleration belts themselves; install an audio sensor on the road between the two deceleration belts, that is, Audio sensors are installed in the detection area;

S2: Use the audio sensor to detect the real-time sound wave waveform formed when the current vehicle passes the two deceleration belts and the road between the two deceleration belts, and record the current vehicle passing the two deceleration belts and the distance between the two deceleration belts at the same time. road time between

S3: Perform the following denoising processing on the detected real-time sound waves:

S3a: Perform filtering and smoothing processing on the real-time waveform, the filtering and smoothing processing is the prior art;

S3b: Determine whether the peak value of the real-time waveform processed by S2a is greater than the threshold value A, the threshold value A is an empirical value, if the peak value of the real-time waveform is greater than the threshold value A, then the real-time waveform is a suspected waveform, and perform the next step; otherwise, return to the step S2, that is, if the peak value of the real-time waveform is less than or equal to the threshold A, return to step S2 for re-detection;

S3c: Calculate the rising edge rate of the suspected waveform. If the rising edge rate is greater than the threshold B, which is an empirical value, then it is determined that the suspected waveform is when the vehicle passes the two speed bumps and the road between the two speed bumps Formed acoustic waveform, otherwise return to step S2, that is, if the rising edge rate is greater than threshold B and less than or equal to threshold A, then return to step S2 for re-detection;

The calculation method of the rising edge rate of the suspected waveform is as follows:

1) Each suspected waveform has at least two peak groups (one peak group will be generated every time a wheel passes through a speed bump, if the vehicle has n rows of wheels, then there will be n peak groups in each peak group, because The suspected waveform acquired may have noise interference, so there may be multiple peak groups).

2) Determine the trough adjacent to the first peak in the first or second peak group, and select the rise between the trough and the first peak (the first peak in the first or second peak group) curve segment, and then randomly select a section in the rising curve segment as the calculation segment, and use the formula (a) to calculate the rising edge rate:

Wherein the ordinate of the two endpoints of the calculation segment is amplitude f, and the abscissa of the two endpoints of the calculation segment is time t.

During specific implementation, it is best to select the part of the rising curve section from the trough to the first peak that is in the 10%-90% (rising time, which can reflect the rapidity of waveform change) part of the rising curve section as the calculation section, and the determined rising edge rate is faster. precise.

Or determine the trough adjacent to the last peak in the first or second peak group, select the descending curve segment between the last peak and the trough (the last peak in the first or second peak group), and then Randomly select a section in the descending curve section as the calculation section, and use the rising edge rate of formula (a).

During specific implementation, it is preferable to select the part of the descending curve segment that is within 25%-75% of the descending curve segment from the trough to the last peak as the calculation segment, so that the determined rising edge rate is more accurate. S4: Note that the sound wave waveform determined in step S3c has two peak groups, wherein the first wave peak group is generated when the wheels of the current vehicle pass through the first isolation zone, and the first wave peak in the first wave peak group is the current vehicle The first row of wheels (that is, the front wheel of the vehicle) is generated when the first row of wheels (that is, the front wheel of the vehicle) passes through the first isolation zone, and the second peak in the first peak group is generated when the second row of wheels of the current vehicle passes through the first isolation zone, and so on , the i-th peak in the first peak group is generated when the i-th row of wheels of the current vehicle passes through the first barrier; the second peak group is generated when the wheels of the current vehicle pass through the second barrier. The first peak in the second peak group is generated when the first row of wheels of the current vehicle (ie, the front wheel of the vehicle) passes through the second barrier, and the second peak in the second peak group is the second row of wheels of the current vehicle Generated when passing through the second isolation zone, and so on, the j-th peak in the second peak group is generated when the j-th row of wheels of the current vehicle passes through the second isolation zone;

The time point corresponding to the i-th peak in the first peak group is recorded as The time point corresponding to the jth peak in the second peak group is recorded as The vehicle speed is calculated according to the formula (1):

$\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; L</span>}}{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2</span>}<span\; class="notranslate">\; ,</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; .</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; D.</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$

Among them, v represents the speed of the vehicle between the two speed reduction belts, L represents the distance between the two speed reduction belts, and D represents the total number of rows of wheels.

As an optimization, the method of the present invention can further determine the classification of the vehicle, the steps are as follows:

1) Calculate the wheelbase of two adjacent rows of wheels according to formula (2):

${\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; L</span>}}{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; \&Center\; Dot;</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$

Where s _k represents the wheelbase of two adjacent rows of wheels;

2) Calculate the total wheelbase s of the vehicle according to formula (3):

$\mathrm{<span\; class="notranslate">\; the\; s</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; D.</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; the\; s</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$

3) According to the total wheelbase s of the vehicle, the classification of the current vehicle can be obtained by comparing the wheelbase information of the vehicle classification (the wheelbase information of the vehicle classification is the existing public data).

As an optimization, the audio sensor is located in the middle of the two barriers.

Combining with Figure 1, assuming that the sound propagation time at L/2 is t ₀ , the vehicle speed can be calculated according to formula (4):

$\mathrm{<span\; class="notranslate">\; v</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; L</span>}}{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; 0</span>}}<span\; class="notranslate">\; )</span>}<span\; class="notranslate">\; =</span>\frac{\mathrm{<span\; class="notranslate">\; L</span>}}{{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; j</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; t</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}^{\mathrm{<span\; class="notranslate">\; 1</span>}}}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; j</span>}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 4</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ;</span>$

Equation (4) is the same as Equation (1), but the position setting of the sensor can overcome the time delay problem caused by a single sensor and improve the detection accuracy.

Finally, it is noted that the above embodiments are only used to illustrate the technical solutions of the present invention without limitation. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be carried out Modifications or equivalent replacements without departing from the spirit and scope of the technical solution of the present invention shall be covered by the claims of the present invention.

Claims (3)

1., based on a vehicle parameter recognition methods for deceleration strip phonetic analysis, it is characterized in that, comprise the steps:

S1: the road between two deceleration strips installs audio sensor;

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

S3: obtain being that real-time sound wave carries out following denoising to detection:

S3a: Filtering and smoothing process is carried out to described real-time waveform;

S3b: judge whether the crest value of the real-time waveform after S2a process is greater than threshold value A, threshold value A is empirical value, if the crest value of real-time waveform is greater than threshold value A, then this real-time waveform is doubtful waveform, and performs next step; Otherwise return step S2;

S3c: the rising edge speed calculating doubtful waveform, if rising edge speed is greater than threshold value B, threshold value B is empirical value, then determine this doubtful waveform be described vehicle cross between these two deceleration strips and two deceleration strips road time the acoustic waveform that formed, otherwise return step S2;

S4: the acoustic waveform determined through step S3c has two crest groups, the time point that in first crest group, i-th crest is corresponding is designated as the time point that in second crest group, a jth crest is corresponding is designated as the speed of a motor vehicle is calculated according to formula (1):

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

Wherein v represents the speed of a motor vehicle of vehicle between two deceleration strips, and L represents the distance between two deceleration strips, and D represents total row of wheel.

2., as claimed in claim 1 based on the vehicle parameter recognition methods of deceleration strip phonetic analysis, it is characterized in that, determine the classification of vehicle, step is as follows:

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

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

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

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

$s=\underset{k=1}{\overset{D-1}{\mathrm{\&Sigma;}}}{s}_k---\left(3\right);$

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

3., as claimed in claim 1 or 2 based on the vehicle parameter recognition methods of deceleration strip phonetic analysis, it is characterized in that, described audio sensor is in the centre position of two isolation strip.