Method, device, terminal and storage medium for measuring ecological noise source

Abstract

The present application provides a method, device, terminal and storage medium for measuring ecological noise sources, which are applicable to the fields of data processing and ecological environment technology. The method includes: obtaining the acoustic and psychoacoustic parameters of the ecological noise source to be measured; obtaining the corresponding target noise level according to the acoustic and psychoacoustic parameters and the preset ecological noise source identification model; obtaining the perceived decibel value and the absolute sound level decibel value corresponding to the target noise level within the preset time length; determining the noise value of the ecological noise source to be measured according to the perceived decibel value and the absolute sound level decibel value. This scheme analyzes the ecological properties of environmental sound, quantitatively measures the ecological noise source, and distinguishes the noise impact of different sound sources. The measured noise value can reflect the noise level of different environmental sound sources under ecological environment conditions. The obtained noise measurement value accurately identifies the environmental noise level, and solves the problem of inaccurate noise monitoring and excessive natural sound caused by environmental noise measurement not reflecting the noise level of the noise source.

Description

Technical Field

[0001] The present application belongs to the field of data processing and ecological environment technology, and in particular, relates to a measurement method, device, terminal and storage medium for ecological noise sources. Background Art

[0002] Noise control based on soundscape believes that noise is the physical and mental stress caused by the loud sounds that people don't want to hear and the auditory irritation caused by the bad environment. Noise control based on soundscape emphasizes the control of indicators that cause auditory irritation. Based on the noise control concept of soundscape, the International Organization for Standardization published the soundscape standard Acoustics-Soundscape (ISO 12913-1:2014) in 2014, which defines the soundscape content, measurement standards, and data analysis requirements. In addition to using equivalent sound level as a noise monitoring indicator, the international community also takes into account noise source indicators. Based on the ISO12913 soundscape standard, it fills the problem of missing noise sources in the original noise control and clarifies that the noise source data needs to be measured and analyzed in the acoustic environment monitoring. However, in general, it only clarifies the location type and sound source type of the sound source, which is a "one-issue-one-discussion" noise indicator. Based on the concept of soundscape, the European Union proposed an environmental noise law to protect quiet areas for its member states. The law clearly points out that natural sound sources have positive significance for human health and should be protected as environmental resources. As a result, countries have started to conduct soundscape surveys in ecological and environmental areas, and the results of the surveys are an important component of environmental noise control. Although European and American countries have carried out measurements of environmental sound sources, they are unable to achieve real-time monitoring, and basically adopt a "one-on-one discussion" approach to noise control based on specific circumstances. This management method is still feasible for developed countries without much construction, but it is not applicable to the rapid urbanization development in my country, and cannot provide accurate data for controlling the increasing noise.

[0003] The soundscape index requires that noise sources be considered, but the environmental noise source monitoring method is very immature and requires sound source identification. Existing sound source identification technology mainly identifies sound signals, and sound signal recognition is suitable for identifying a single sound source, whether it is a steady-state sound source, a non-steady-state sound source or a moving sound source. In addition, the sound signal recognition technology has poor resolution for low-frequency sounds and cannot solve the problem of back-to-back interference sources. Therefore, current sound recognition technology is not suitable for identifying the irritability of environmental sound sources, because the identification of a single sound source is not suitable for identifying environmental sound sources, let alone judging the noisiness of different environmental sound sources. Noise monitoring is not accurate and cannot be directly applied to noise source monitoring work. Summary of the invention

[0004] In view of this, the embodiments of the present application provide a method, device, terminal and storage medium for identifying ecological noise sources, which can solve the problems that the current sound recognition technology is not suitable for identifying the irritability of environmental sound sources, the noise monitoring is not accurate, and cannot be directly applied in noise source monitoring work.

[0005] A first aspect of an embodiment of the present application provides a method for identifying an ecological noise source, comprising:

[0006] Obtain the ecological noise source to be measured;

[0007] Acquiring acoustic and psychoacoustic parameters of the ecological noise source to be measured;

[0008] According to the acoustic and psychoacoustic parameters of the ecological noise source to be measured and the preset ecological noise source identification model, the target noise level of the ecological noise source to be measured output by the ecological noise source identification model is obtained; the preset ecological noise source identification model is obtained based on a plurality of ecological noise source samples and the target noise level of each ecological noise source sample is trained, and the target noise level of each ecological noise source sample is obtained based on cluster analysis of each ecological noise source sample;

[0009] Obtaining a perceived decibel value and an absolute sound level decibel value corresponding to a target noise level of the ecological noise source to be measured within a first preset time period;

[0010] The noise value of the ecological noise source to be measured is determined according to the perceived decibel value and the absolute sound level decibel value corresponding to the target noise level of the ecological noise source to be measured.

[0011] A second aspect of an embodiment of the present application provides an ecological noise source identification device, including:

[0012] An ecological noise source acquisition module, used to acquire the ecological noise source to be measured;

[0013] An acoustic parameter acquisition module, used to acquire the acoustic and psychoacoustic parameters of the ecological noise source to be measured;

[0014] a noise level determination module, configured to obtain, according to the acoustic and psychoacoustic parameters of the ecological noise source to be measured and a preset ecological noise source identification model, a target noise level of the ecological noise source to be measured output by the ecological noise source identification model; the preset ecological noise source identification model is obtained by training based on a plurality of ecological noise source samples and the target noise level of each ecological noise source sample, and the target noise level of each ecological noise source sample is obtained based on cluster analysis of each ecological noise source sample;

[0015] A decibel value acquisition module is used to acquire a perceived decibel value and an absolute sound level decibel value corresponding to a target noise level of the ecological noise source to be measured within a first preset time period; and

[0016] The noise value determination module is used to determine the noise value of the ecological noise source to be measured according to the perceived decibel value and the absolute sound level decibel value corresponding to the target noise level of the ecological noise source to be measured.

[0017] A third aspect of an embodiment of the present application provides a terminal device, which includes a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the steps of the ecological noise source measurement method described in the first aspect are implemented.

[0018] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, which stores a computer program, and is characterized in that when the computer program is executed by a processor, the steps of the method for measuring ecological noise sources as described in the first aspect are implemented.

[0019] A fifth aspect of the embodiments of the present application provides a computer program product. When the computer program product is run on a terminal, the terminal executes the steps of the ecological noise source measurement method described in the first aspect.

[0020] Compared with the prior art, the embodiments of the present application have the following beneficial effects: the present application obtains the target noise level of the ecological noise source to be measured through the acoustic and psychoacoustic parameters of the ecological noise source to be measured and the preset ecological noise source identification model, and then the noise value of the ecological noise source to be measured is jointly determined by the perceived decibel value corresponding to the target noise level of the ecological noise source to be measured and the absolute sound level decibel value. The present application analyzes the ecological properties of environmental sound, quantitatively measures the ecological noise source, and distinguishes the noise influence of different sound sources. The measured noise value can reflect the noise level of different environmental sound sources under ecological environment conditions. The obtained noise measurement value accurately identifies the environmental sound noise level and belongs to real noise. It solves the current problem that the environmental noise measurement does not reflect the noise characteristics of the noise source under the construction of the ecological environment, resulting in inaccurate noise monitoring, and the problem of excessive natural sound in the ecological environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0022] Figure 1 It is an application environment diagram of the ecological noise source measurement method provided in the embodiment of the present application;

[0023] Figure 2 is a flow chart of a method for measuring an ecological noise source provided in an embodiment of the present application;

[0024] Figure 3 This is a schematic diagram of noise source level classification provided by an embodiment of the present application;

[0025] Figure 4 is a flow chart of cluster analysis of various ecological noise source samples provided in an embodiment of the present application;

[0026] Figure 5 is a graph showing the relationship between the sum of squared errors and the number of clusters provided in an embodiment of the present application;

[0027] Figure 6 is a flow chart of generating a preset ecological noise source identification model provided in an embodiment of the present application;

[0028] Figure 7 is a flow chart of an optimized ecological noise source measurement method provided in an embodiment of the present application;

[0029] Figure 8 It is a structural block diagram of the device for measuring ecological noise sources provided in an embodiment of the present application;

[0030] Fig. 9 It is a schematic diagram of a terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0031] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0032] At present, with the construction of ecological cities, combined with the phenomenon of excessive noise caused by natural sounds, the current noise environment measurement technology has the following defects: 1) As cities develop towards green and ecological directions, more and more natural sounds and popular life sounds appear in cities. The equivalent continuous A sound level Leq is used as a noise level monitoring method. It does not distinguish the influence of different sound sources, which is particularly unsuitable for the situation where there are many natural sounds in the ecological city environment; 2) The current sound recognition technology mainly corresponds to a single sound source. In the case of multiple sound sources, the sound source with insignificant audio signal characteristics is difficult to identify; 3) At present, the noise control based on soundscape has no quantitative indicators for the measurement of sound sources, and only gives the sound category, so it is difficult to achieve automatic measurement. For the ecological noise environment, real-time monitoring cannot be achieved; 4) The perception and recognition of "noise" based on a single sound source is completely different from the perception and recognition of "noise" containing multiple sound sources in ecological noise. The latter has multiple sound source components, and the noise levels of various sound sources are different and difficult to quantify, so it is difficult to identify it. There is no corresponding recognition technology at present.

[0033] In this case, the research work on the perception of irritability through environmental noise sources provides a theoretical basis for the measurement of environmental noise sources in eco-cities. This application proposes four types of environmental noise sources by investigating the correlation between various noise sources and auditory irritability perception in eco-cities. Based on the amount of natural ecological sounds contained in each type of sound source, the irritability perception from type 1 to type 4 is getting higher and higher. Among them, the irritability brought to people by the third type of environmental noise source is consistent with the corresponding equivalent sound level, that is: the noise control standards currently implemented are actually for the third type of environmental noise sources, and people’s actual irritability for the first and second types of environmental noise sources is lower than the control standard, and the actual irritability for the fourth type of environmental noise source is higher than the control standard.

[0034] In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.

[0035] Figure 1 An application environment diagram of an ecological noise source measurement method provided in an embodiment of the present application, namely, an outdoor ecological noise source measurement device, such as Figure 1 As shown, in this application environment, an outdoor integrated electrical cabinet is built, and the ecological noise source identification model and the ecological noise source measurement procedure are combined with the existing measurement equipment. Figure 1 (a) shows the external schematic diagram of the outdoor integrated electrical cabinet. Figure 1(b) shows the internal schematic diagram of the outdoor integrated electric cabinet, which mainly includes three parts: acoustic signal analysis equipment, ecological noise value calculation module and data transmission equipment. The current common outdoor sound collection equipment on the market is connected to the outdoor integrated electric cabinet, and the outdoor sound collection equipment is used to collect environmental sound; based on the connection of the front-end sound collector, the sound signal data transmission line, and the multi-channel sound data acquisition analyzer, the sound acoustic and psychoacoustic data that can analyze the sound source category are obtained; after obtaining the basic acoustic data and psychoacoustic data, the ecological noise value calculation module is combined to determine the noise source category, that is, the target noise level, and the noise value of the ecological noise source is calculated by the noise source category; finally, the acoustic data obtained by the front-end analysis can be remotely transmitted to the back-end server through the data transmission device. This technology can change the existing noise measurement equipment. By implanting the calculation model of sound source identification into the existing measuring instrument, it can be more convenient to carry out noise monitoring based on ecological noise source identification, adapt to the development of ecological cities, and scientifically prevent and control urban noise pollution.

[0036] like Figure 2 As shown, in one embodiment, a method for measuring ecological noise sources is proposed, the method comprising the following steps:

[0037] Step S201, obtaining the ecological noise source to be measured.

[0038] In the embodiment of the present application, the ecological noise source refers to the environmental sound. A subjective irritation perception experiment is conducted on various environmental sounds that may exist in an ecological city under the same sound level conditions. The environmental sound can be graded for subjective irritation perception according to the three semantic categories of sounds it contains: natural sound, life sound, mechanical sound, and mixed sound. Figure 3 As shown, it can be divided into 9 levels.

[0039] Step S202: Acquire the acoustic and psychoacoustic parameters of the ecological noise source to be measured.

[0040] In the embodiment of the present application, the degree of noise that people feel from different environmental sound sources is closely related to acoustic parameters. Acoustic and psychoacoustic parameters refer to the frequency domain characteristics, time domain characteristics, amplitude characteristics, and psychoacoustic characteristics of environmental sound. The above characteristics are factors that affect people's perception of environmental sound sources, and are also indicators for judging whether environmental sound sources are noisy. Specifically, the acoustic and psychoacoustic parameters of the ecological noise source to be measured include acoustic feature quantities and statistical characteristics of the acoustic feature quantities based on time changes; the acoustic feature quantities include one or more acoustic feature quantities in sound level, loudness, roughness, sharpness, jitter, and pitch; the statistical characteristics include one or more statistical characteristics in mean, maximum value, standard deviation, and percentile.

[0041] Step S203: obtaining the target noise level of the ecological noise source to be measured output by the ecological noise source identification model according to the acoustic and psychoacoustic parameters of the ecological noise source to be measured and the preset ecological noise source identification model.

[0042] In an embodiment of the present application, the preset ecological noise source identification model is trained based on multiple ecological noise source samples and the target noise level of each ecological noise source sample. The target noise level of each ecological noise source sample is obtained based on cluster analysis of the each ecological noise source sample.

[0043] Optionally, in the embodiment of the present application, Figure 4 As shown, obtaining the target noise level of each ecological noise source sample based on cluster analysis of each ecological noise source sample includes the following steps:

[0044] Step S401, obtaining acoustic and psychoacoustic parameters of each ecological noise source sample.

[0045] Step S402, according to the acoustic and psychoacoustic parameters of each of the ecological noise source samples, classify and merge similar sound clips of each of the ecological noise source samples to obtain multiple groups of sound clips; each group of sound clips contains sound clips of similar noise levels.

[0046] In an embodiment of the present application, more than 60,000 ecological noise source sample data measured in the field are taken as an example, and similar sound fragments are classified and merged by classifying the more than 60,000 ecological noise source sample data measured in the field. The method is based on their acoustic parameters. First, the sound fragments of the same time are compared in terms of their equivalent sound level, L10, L50, L90 and other sound levels, and the fragments without significant differences are classified, and typical fragments are selected from each category. Then compare the psychoacoustic parameters, sound pressure level, frequency and other parameters of the typical fragments, and also classify the fragments without significant differences, so that more than 60,000 sound source fragments are summarized into more than 200 sound source fragment samples. Then conduct a listening experiment on these samples, merge the samples with the same sound semantics, and finally obtain 44 samples for laboratory subjective evaluation experiments.

[0047] Step S403: obtaining a quietness evaluation value corresponding to each group of sound clips.

[0048] Optionally, in an embodiment of the present application, obtaining the quietness evaluation value of each group of sound clips includes: cutting each sound clip contained in each group of sound clips according to a second preset time length to obtain a plurality of sub-sound clips; adjusting the sound pressure level of the middle clip of each sub-sound clip to a preset value to obtain an adjusted sound clip; and obtaining the quietness evaluation value of the adjusted sound clip.

[0049] Specifically, taking the above as an example, the 44 sound clips obtained from field measurements were cut into 45-second segments, and the middle segments were taken to adjust their sound pressure levels to 65 decibels. At this time, people's evaluation of the quietness of different sound source segments will come from the difference in sound categories, rather than being affected by the sound level. All the segments were subjected to the subjective evaluation experiment of noise quietness, and a total of more than 13,000 sound quietness evaluation values ​​were obtained.

[0050] Step S404: Obtain the sum of square errors of the quietness evaluation values ​​corresponding to each group of sound clips.

[0051] Step S405, obtaining the variation trend of the sum of square errors of the quietness evaluation values ​​corresponding to each group of sound clips as the number of noise levels of the ecological noise source samples increases, and determining the target noise level corresponding to each of the ecological noise source samples.

[0052] In an embodiment of the present application, the above-mentioned more than 13,000 quietness evaluations are clustered according to the Kmeans clustering algorithm in the statistical method, and the elbow method is used to determine the irritability level (K) of the noise source, that is, the number of target noise levels. The core of the elbow method is SSE (sum of the squared errors). The principle is that as the number of clusters K increases, the sample division becomes more refined, and SSE will decrease. When K is less than a reasonable number of clusters, the increase in K will greatly increase the degree of aggregation of each cluster, and the SSE will decrease significantly. When K reaches a reasonable number of clusters, if K is increased again, the SSE will decrease sharply, and then it will tend to be flat as the K value increases. The relationship between SSE and K is like an elbow, so a reasonable K should be at the place where the amplitude changes on both sides, that is, the elbow. Figure 5 As shown. The elbow method in statistics was used to judge the error sum of squares, and it was found that the error sum of squares decreased rapidly when the number of clusters was less than 4 or 5, while the SSE decreased slowly when the number of clusters was more than 4 or 5. Further experiments were conducted by using different numbers of quietness evaluations and taking the method of hearing verification. It was found that dividing different sound sources into 4 categories is more in line with people's irritability perception of different environmental sounds. It was determined that based on people's auditory quietness / irritability perception, Shenzhen's urban environmental sound has 4 major categories, which are defined as noise sources 1, 2, 3, and 4 according to the different irritability perceptions they bring to people.

[0053] In the embodiment of the present application, noise source 1 brings the lowest sense of irritation to people, and 4 brings the highest sense of irritation to people. Through listening experiments on 44 experimental sound samples, it was found that the main sounds contained in the first type of noise source are natural sounds, including waves, flowing water, birds and insects, and their mixtures. When the sound pressure level is the same at 65 decibels, people's quietness evaluation is 5.48; the second category is mainly mixed sounds with obvious natural sounds and life sounds with clear semantics. The evaluation of this type of sound source is also relatively high, mainly between 4-5; the third category is mainly mixed sounds with more life sounds. Although natural sounds are also involved, the impact is weaker. The evaluation value of this type of sound source is mainly around 3.58; the fourth level is mainly mechanical sounds and sound source types dominated by mechanical sounds. The quietness of this type of sound source is significantly worse than other types, and the evaluation values ​​are all around 2.51.

[0054] In the embodiments of the present application, as the components of ecological noise sources in different geographical conditions are different, the noise levels of the ecological noise sources obtained will also be different. Based on the existing soundscape research results and a large amount of environmental sound field survey data, this application divides the noise sources into four categories according to the components of the ecological noise sources through cluster calculation, and is not used to limit the scope of this application.

[0055] Optionally, in the embodiment of the present application, Figure 6 As shown, the preset ecological noise source identification model is obtained based on multiple ecological noise source samples and the target noise level training of each ecological noise source sample, including the following steps:

[0056] Step S601, obtaining multiple ecological noise source samples.

[0057] Step S602, obtaining acoustic and psychoacoustic parameters of each of the ecological noise source samples;

[0058] In the embodiment of the present application, the above characteristics of the ambient sound used in the subjective evaluation experiment are extracted from each of the ecological noise source samples, and the artificial intelligence model learning is applied to identify four different types of ecological environmental noise sources. The following is the time domain change of the ambient sound amplitude, representing the loudness characteristics:

[0059]

[0060] Where: E TQ is the stimulus corresponding to the hearing threshold in quiet conditions; E is the stimulus corresponding to the calculated sound; E0 is the reference sound intensity; Corresponding incentives.

[0061] The psychoacoustic indicators sharpness, roughness, and tonality most accurately reflect people's psychological feelings about different types of sounds. The calculation formula for sharpness is as follows:

[0062]

[0063] Where: g(z) is the weighting coefficient,

[0064] The calculation formula of roughness is as follows:

[0065]

[0066] Where: ΔL E is the amplitude of sound pressure change in each critical frequency band; f mod is the modulation frequency; f0 is the modulation base frequency, f0=70Hz.

[0067] The tonality calculation formula is as follows:

[0068]

[0069] Where: W1(Δz i ) is the relationship between the ith single frequency component and the critical frequency band difference; W2(f1) is the relationship between the ith single frequency component and the frequency; W3(ΔL i ) represents the sound level surplus effect of the i-th single frequency component.

[0070] The time domain characteristics of ambient sound are reflected by indicators such as fluctuation strength and the change of sound level over time. The calculation formula is as follows:

[0071]

[0072] Where: N′ max ,N′ min is the maximum and minimum specific loudness in each critical frequency band; f mod is the modulation frequency; f0 is the modulation base frequency, f0=70Hz.

[0073] L w , Lesm, Sslop2,, Tesm, Tphimax, and crz are all autocorrelation function characteristics of sound level.

[0074] For the above sound characteristics such as amplitude, frequency domain, and psychoacoustic quantity, we further calculate their statistics based on time domain changes: mean, maximum value, standard deviation, 10th percentile (p10 ), 25th percentile (p 25 ), 50 percentile (p 50 ), 75th percentile (p 75 ), 90th percentile (p 90 ), the difference between the 10th percentile and the 90th percentile (p 10 -p 90 ), 54-dimensional related factors were extracted, as shown in Table 1. Their relationship with the judgment of four types of ecological noise sources was studied to determine the main factors affecting the categories of ecological noise sources and provide a scientific basis for artificial intelligence machine learning.

[0075] Table 1. Feature extraction of sound samples

[0076]

[0077]

[0078] Step S603, inputting the acoustic and psychoacoustic parameters of each ecological noise source sample into a neural network model containing variable parameters, and obtaining the corresponding response noise level of each ecological noise source sample output by the neural network model containing variable parameters.

[0079] In an embodiment of the present application, the neural network model structure may include an input layer, multiple convolutional layers, multiple pooling layers, at least one fully connected layer, and an output layer, wherein the multiple convolutional layers and the multiple fully connected layers have variable parameters. When the variable parameters in the multiple convolutional layers and the multiple fully connected layers change, the output response noise level is different for the same acoustic and psychoacoustic parameter input.

[0080] Step S604: Acquire the corresponding response noise level of each ecological noise source sample and each loss difference between the response noise level and the corresponding target noise level of each ecological noise source sample.

[0081] Step S605, determining whether the loss differences of the ecological noise source samples meet a preset condition.

[0082] In an embodiment of the present application, when it is determined that the loss differences of the various ecological noise source samples do not meet the preset conditions, it indicates that the trained ecological noise source identification model has a large error, and step S606 is executed; when it is determined that the loss differences of the various ecological noise source samples meet the preset conditions, it indicates that the trained ecological noise source identification model is relatively stable, and step S607 is executed.

[0083] Step S606, adjusting the variable parameters in the neural network model.

[0084] In an embodiment of the present application, the variable parameters in the convolutional neural network model can be adjusted according to the back propagation algorithm or the gradient descent algorithm, but after adjusting the variable parameters, return to step S603 to recalculate the loss difference.

[0085] Step S607: determine the current neural network model containing variable parameters as an ecological noise source identification model.

[0086] In an embodiment of the present application, in order to obtain an ecological noise source identification model, that is, to obtain the implicit relationship between acoustic parameters and target noise levels, it is necessary to take multiple acoustic parameters-target noise levels as samples, calculate the difference between the response noise level corresponding to the input acoustic parameters and the target noise level under the current form of the neural network model, and continuously feedback and adjust the variable parameters in the neural network model until, under a certain neural network model, the difference between the response noise level corresponding to the input acoustic parameters and the target noise level meets the preset conditions. At this time, the neural network model is the required preset ecological noise source identification model.

[0087] Step S204: obtaining a perceived decibel value and an absolute sound level decibel value corresponding to a target noise level of the ecological noise source to be measured within a first preset time period.

[0088] In an embodiment of the present application, the ecological noise source identification model can judge the noise level of the ambient noise based on the amplitude, time domain, frequency domain, and psychoacoustic characteristics of the ambient sound. At present, the sound level is a physical indicator that indicates the effect of noise on people's irritability perception (the higher the sound level, the noisier the environment in which people live). Therefore, it is necessary to convert the noise level of the ecological noise source into a sound level value in order to be connected with the existing noise standard measurement. This application will measure the noise in the ecological environment more accurately by adding the noise level value of the ecological noise source to the existing noise measurement. In the auditory subjective evaluation experimental environment, the "irritability" of the four types of ecological noise sources at the same sound level is marked with the sound level size. The "noise" of this type of ecological noise source whose sound level value is consistent with the noise perception is represented by 0dB, and its noise index is the sound level of the ambient sound, which is identified as a standard ecological noise source. When the "noise" of a certain type of ecological noise source is higher than that of the standard ecological noise source at the same sound level, the "noise" brought by the sound source is greater than 0dB, which is a positive value, and this type of noise source makes people more irritated; on the contrary, the "noise" brought by the sound source is less than 0dB, which is a negative value, and this type of noise source makes people quieter. The noisier / quieter values ​​brought by different ecological noise sources are measured with corresponding sound level values. The measurement method is through a subjective evaluation experiment, asking the experimenter to adjust the sound level to the noise level value he / she thinks (that is, adjust the sound level to the noise value he / she thinks).

[0089] Step S205: determining the noise value of the ecological noise source to be measured according to the perceived decibel value and the absolute sound level decibel value corresponding to the target noise level of the ecological noise source to be measured.

[0090] In the embodiment of the present application, according to the ambient sound level calculation formula combined with the noise level assignment of the "irritability" of the four types of noise sources, the noise level calculation formula of the ecological noise source can be listed as follows:

[0091]

[0092] Where: L A is the instantaneous A sound level at time t; T is the specified measurement time period; x is the "noise level" value corresponding to the four types of noise sources. It represents the sound pressure level measured within a specified time, which is added to the noise level value x corresponding to the ecological noise source (that is, the "noise degree" corresponding to each sound source), and the result is used to reflect people's actual irritation perception of the ecological noise source.

[0093] This application obtains the target noise level of the ecological noise source to be measured through the acoustic and psychoacoustic parameters of the ecological noise source to be measured and the preset ecological noise source identification model, and then determines the noise value of the ecological noise source to be measured by the perceived decibel value corresponding to the target noise level of the ecological noise source to be measured and the absolute sound level decibel value. This application analyzes the ecological properties of environmental sound, quantitatively measures ecological noise sources, and distinguishes the noise impact of different sound sources. The measured noise value can reflect the noise level of different environmental sound sources under ecological environmental conditions. The obtained noise measurement value accurately identifies the environmental sound noise level and belongs to real noise. It solves the current problem that the environmental noise measurement does not reflect the noise characteristics of the noise source under the construction of the ecological environment, resulting in inaccurate noise monitoring, and the problem of excessive natural sound in the ecological environment.

[0094] To correctly measure the irritability of environmental noise to people, it is necessary to measure the environmental noise source index. Since the environmental noise source index reflects people's auditory perception, artificial intelligence technology will play a key role. This application uses artificial intelligence technology to identify 4 types of environmental sound sources, and weights the sound level value equivalent to the irritability perception of the environmental noise source with the equivalent sound level of the environmental sound as the ecological noise source measurement value. This value can truly reflect the noise pollution in the ecological city environment and meet the requirements for measuring the ecological city noise environment. In response to the specific requirements of my country's green ecological development, the invention of a noise source monitoring technology that meets my country's national conditions can not only ensure the accuracy of environmental noise monitoring, but also solve the problem that there are currently no quantitative indicators for noise source monitoring in the world, which has international leading significance.

[0095] As a preferred embodiment of the present application, Figure 7 As shown, after step S202, the following further includes:

[0096] Step S701 , performing dimensionality reduction processing on the acoustic and psychoacoustic parameters of the ecological noise source to be measured to obtain acoustic and psychoacoustic parameters after dimensionality reduction.

[0097] In an embodiment of the present application, the acoustic and psychoacoustic parameters after dimensionality reduction include acoustic feature quantities and statistical characteristics of the acoustic feature quantities based on time changes; the acoustic feature quantities include one or more acoustic feature quantities of sound level, loudness, roughness, sharpness, jitter and pitch; the statistical characteristics include one or more statistical characteristics of mean, standard deviation, 25th percentile, 50th percentile, 75th percentile, 90th percentile and the difference between the 10th percentile and the 90th percentile.

[0098] Specifically, four types of ecological environmental noise sources with different irritability perceptions were identified. First, the sound level and psychoacoustic features were extracted for each sound clip, and then nine statistical features of six features were calculated: mean, maximum, standard deviation, 10th percentile, 25th percentile, 50th percentile, 75th percentile, 90th percentile, and the difference between the 10th percentile and the 90th percentile, and a total of 54 dimensions of relevant features were extracted. The 54-dimensional factors affecting the judgment of ecological noise sources were reduced in dimension through correlation statistical analysis, and the features significantly correlated with the subjective irritability evaluation were refined to determine the factors affecting the judgment of noise source categories as the construction conditions of the input variables of the machine learning model. Based on the theoretical model of ecological noise source classification in the early stage, the learning effects of artificial neural network models, decision tree models, etc. were studied.

[0099] Before machine learning, the acoustic software is used to extract the acoustic features and their time-varying statistical values ​​of all the learned sound source segments (including 4 types of noise sources), a total of 42 dimensions, namely L m , L s , L 25 , L 50 , L 75 , L 10 -L 90 , Ld m , Ld s , Ld 25 , Ld 50 , Ld 75 , Ld 10 -Ld 90 , R m , R s , R 25 , R 50 , R 75 , R 10 -R 90, S m , S s , S 25 , S 50 , S 75 , S 10 -S 90 、F m 、F s 、F 25 、F 50 、F 75 、F 10 -F 90 , T m , T s , T 25 , T 50 , T 75 , T 10 -T 90 ,t el , l w , l esm 、s slop2 ,、t esm ,t phimax , crz. For different combinations of input variables, the calculation results of different artificial neural network and decision tree model structures were compared, and the most accurate and efficient model was selected as the ecological noise source identification model to judge the four types of ecological noise sources. Then, the model judgment results were verified by using external data of model learning, and the accuracy of the model was calibrated. The model with a prediction accuracy of more than 0.85 was selected as the model that can be used for ecological noise source measurement, namely, the ecological noise source identification model.

[0100] The ecological noise source measurement method provided in the embodiment of the present application reduces the dimension of the 54-dimensional factors that affect the judgment of the ecological noise source through correlation statistical analysis, and extracts the features that are significantly correlated with the subjective irritability evaluation to determine the factors that affect the judgment of the noise source category. As the construction conditions of the input variables of the machine learning model, an accurate and efficient ecological noise source identification model can be obtained, thereby further improving the measurement accuracy of the ecological noise source.

[0101] Corresponding to the method of the above embodiment, Figure 8 A structural block diagram of an ecological noise source measurement device provided in an embodiment of the present application is shown. For ease of explanation, only the parts related to the embodiment of the present application are shown. Figure 8 The exemplary ecological noise source measurement device may be an execution subject of the ecological noise source measurement method provided in the aforementioned embodiment.

[0102] Reference Figure 8 The measuring device of the ecological noise source comprises:

[0103] The ecological noise source acquisition module 810 is used to acquire the ecological noise source to be measured;

[0104] An acoustic parameter acquisition module 820 is used to acquire acoustic and psychoacoustic parameters of the ecological noise source to be measured;

[0105] The noise level determination module 830 is used to obtain the target noise level of the ecological noise source to be measured output by the ecological noise source identification model according to the acoustic and psychoacoustic parameters of the ecological noise source to be measured and the preset ecological noise source identification model; the preset ecological noise source identification model is trained based on multiple ecological noise source samples and the target noise level of each ecological noise source sample, and the target noise level of each ecological noise source sample is obtained based on cluster analysis of each ecological noise source sample;

[0106] A decibel value acquisition module 840 is used to acquire a perceived decibel value and an absolute sound level decibel value corresponding to a target noise level of the ecological noise source to be measured within a first preset time period; and

[0107] The noise value determination module 850 is used to determine the noise value of the ecological noise source to be measured according to the perceived decibel value and the absolute sound level decibel value corresponding to the target noise level of the ecological noise source to be measured.

[0108] The process of each module in the ecological noise source measurement device provided in the embodiment of the present application realizing its own function can be specifically referred to the aforementioned Figure 2 The description of the illustrated embodiment will not be repeated here.

[0109] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0110] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0111] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0112] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0113] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish descriptions, and cannot be understood as indicating or suggesting relative importance. It should also be understood that although the terms "first", "second", etc. are used to describe various elements in some embodiments of the present application in the text, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, the first table can be named as the second table, and similarly, the second table can be named as the first table without departing from the scope of the various described embodiments. The first table and the second table are both tables, but they are not the same table.

[0114] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0115] Fig. 9 Schematic diagram of the structure of a terminal device provided by an embodiment of the present application. Fig. 9 As shown, the terminal device 7 of this embodiment includes: at least one processor 70 ( Fig. 9 Only one is shown in the figure), a memory 71, in which a computer program 72 that can be run on the processor 70 is stored. When the processor 70 executes the computer program 72, the steps in the above-mentioned measurement method of each ecological noise source are implemented, such as Figure 2 Alternatively, when the processor 70 executes the computer program 72, the functions of each module / module in the above-mentioned device embodiments are realized, for example Figure 8 Functions of modules 810 to 850 are shown.

[0116] The terminal device 7 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor 70 and a memory 71. Those skilled in the art will appreciate that Figure 7 It is only an example of the terminal device 7 and does not constitute a limitation on the terminal device 7. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include an input sending device, a network access device, a bus, etc.

[0117] The processor 70 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0118] In some embodiments, the memory 71 may be an internal storage module of the terminal device 7, such as a hard disk or memory of the terminal device 7. The memory 71 may also be an external storage device of the terminal device 7, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device 7. Further, the memory 71 may also include both an internal storage module and an external storage device of the terminal device 7. The memory 71 is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of the computer program. The memory 71 may also be used to temporarily store data that has been sent or is to be sent.

[0119] In addition, each functional module in each embodiment of the present application can be integrated into a processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or software functional modules.

[0120] An embodiment of the present application also provides a terminal device, which includes at least one memory, at least one processor, and a computer program stored in the at least one memory and executable on the at least one processor, wherein when the processor executes the computer program, the terminal device implements the steps in any of the above-mentioned method embodiments.

[0121] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned method embodiments can be implemented.

[0122] An embodiment of the present application provides a computer program product. When the computer program product runs on a terminal device, the terminal device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0123] If the integrated module / module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0124] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0125] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0126] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0127] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for measuring ecological noise sources, characterized in that: include: Obtain the ecological noise source to be measured; Acquiring acoustic and psychoacoustic parameters of the ecological noise source to be measured; According to the acoustic and psychoacoustic parameters of the ecological noise source to be measured and a preset ecological noise source identification model, obtaining a target noise level of the ecological noise source to be measured output by the ecological noise source identification model; The preset ecological noise source identification model is obtained by training based on a plurality of ecological noise source samples and a target noise level of each ecological noise source sample, and the target noise level of each ecological noise source sample is obtained based on cluster analysis of each ecological noise source sample; Obtaining a perceived decibel value and an absolute sound level decibel value corresponding to a target noise level of the ecological noise source to be measured within a first preset time period; The noise value of the ecological noise source to be measured is determined according to the perceived decibel value and the absolute sound level decibel value corresponding to the target noise level of the ecological noise source to be measured.

2. The method for measuring ecological noise sources according to claim 1, characterized in that: The step of obtaining the target noise level of each ecological noise source sample based on cluster analysis of each ecological noise source sample includes: Obtain the acoustic and psychoacoustic parameters of each ecological noise source sample; According to the acoustic and psychoacoustic parameters of each of the ecological noise source samples, similar sound clips of each of the ecological noise source samples are classified and merged to obtain multiple groups of sound clips; each sound clip contained in each group of sound clips is a sound clip of the same noise level; Obtaining a quietness evaluation value corresponding to each group of sound clips; Obtaining the sum of square errors of the quietness evaluation values ​​corresponding to each group of sound clips; The variation trend of the sum of square errors of the quietness evaluation values ​​corresponding to each group of sound clips as the number of noise levels of the ecological noise source samples increases is obtained, and the target noise level corresponding to each of the ecological noise source samples is determined.

3. The method for measuring ecological noise sources according to claim 2, characterized in that: The obtaining of the quietness evaluation value of each group of sound clips includes: Cutting each sound segment contained in each group of sound segments according to a second preset time length to obtain a plurality of sub-sound segments; Adjusting the sound pressure level of the middle segment of each sub-sound segment to a preset value to obtain an adjusted sound segment; Get the quietness rating of the adjusted sound clip.

4. The method for measuring ecological noise sources according to claim 1, characterized in that: The preset ecological noise source identification model is obtained by training based on multiple ecological noise source samples and target noise levels of each ecological noise source sample, including: Obtain multiple samples of ecological noise sources; Acquiring acoustic and psychoacoustic parameters of each of the ecological noise source samples; Inputting the acoustic and psychoacoustic parameters of each ecological noise source sample into a neural network model with variable parameters, and obtaining the corresponding response noise level of each ecological noise source sample output by the neural network model with variable parameters; Acquire the corresponding response noise level of each ecological noise source sample and each loss difference between the response noise level and the corresponding target noise level of each ecological noise source sample; Determine whether the loss differences of the ecological noise source samples meet the preset conditions; When it is determined that the loss differences of the ecological noise source samples do not meet the preset conditions, the variable parameters in the neural network model are adjusted, and the process returns to the step of inputting the acoustic and psychoacoustic parameters of the ecological noise source samples into the neural network model containing variable parameters to obtain the corresponding response noisiness level of the ecological noise source samples output by the neural network model containing variable parameters; When it is determined that the loss differences of the ecological noise source samples meet the preset conditions, the current neural network model containing variable parameters is determined as the ecological noise source identification model.

5. The method for measuring ecological noise sources according to claim 1, characterized in that: After obtaining the acoustic and psychoacoustic parameters of the ecological noise source to be measured, the method further includes: The acoustic and psychoacoustic parameters of the ecological noise source to be measured are subjected to dimensionality reduction processing to obtain acoustic and psychoacoustic parameters after dimensionality reduction.

6. The method for measuring ecological noise sources according to claim 5, characterized in that: The acoustic and psychoacoustic parameters of the ecological noise source to be measured include acoustic feature quantities and statistical characteristics of the acoustic feature quantities based on time changes; the acoustic feature quantities include one or more acoustic feature quantities of sound level, loudness, roughness, sharpness, jitter and pitch; the statistical characteristics include one or more statistical characteristics of mean, maximum value, standard deviation and percentile.

7. The method for measuring ecological noise sources according to claim 5, characterized in that: The acoustic and psychoacoustic parameters after dimensionality reduction include acoustic feature quantities and statistical characteristics of the acoustic feature quantities based on time changes; the acoustic feature quantities include one or more acoustic feature quantities of sound level, loudness, roughness, sharpness, jitter and pitch; the statistical characteristics include one or more statistical characteristics of mean, standard deviation, 25th percentile, 50th percentile, 75th percentile, 90th percentile and the difference between 10th percentile and 90th percentile.

8. A device for measuring ecological noise sources, characterized in that: include: An ecological noise source acquisition module, used to acquire the ecological noise source to be measured; An acoustic parameter acquisition module, used to acquire the acoustic and psychoacoustic parameters of the ecological noise source to be measured; A noisiness level determination module, configured to obtain a target noise level of the ecological noise source to be measured output by the ecological noise source identification model according to the acoustic and psychoacoustic parameters of the ecological noise source to be measured and a preset ecological noise source identification model; The preset ecological noise source identification model is obtained by training based on a plurality of ecological noise source samples and a target noise level of each ecological noise source sample, and the target noise level of each ecological noise source sample is obtained based on cluster analysis of each ecological noise source sample; A decibel value acquisition module, used to acquire a perceived decibel value and an absolute sound level decibel value corresponding to a target noise level of the ecological noise source to be measured within a first preset time period; as well as The noise value determination module is used to determine the noise value of the ecological noise source to be measured according to the perceived decibel value and the absolute sound level decibel value corresponding to the target noise level of the ecological noise source to be measured.

9. A terminal device, characterized in that: The terminal device includes a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the computer program, the steps of the ecological noise source measurement method as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for measuring ecological noise sources according to any one of claims 1 to 7 are implemented.

Images