A method and device for evaluating the emotional impact of a road acoustic environment

Abstract

The application discloses a road sound environment emotion influence evaluation method and device, and relates to the technical field of environment evaluation, and comprises the following steps: in the multi-class urban road sound environment function area sampling, determining the effective dominant audio and the dominant sound pressure level of a plurality of road dominant sound source types, and classifying the effective dominant audio and the dominant sound pressure level into a plurality of road noise types; combining the effective dominant audio to build a questionnaire platform and collecting subjective scores of a plurality of subjective indexes; collecting electroencephalogram signals according to the effective dominant audio of each road noise type; determining subjective key principal components according to the subjective scores of each road noise type; extracting a plurality of spectral features of each electroencephalogram signal, and determining objective key principal components according to the spectral features of each road noise type; and respectively adopting the subjective key principal components, the objective key principal components, the effective dominant audio and the dominant sound pressure level to comprehensively determine an emotion comprehensive index of each road noise type. The application is helpful to more comprehensively and accurately reveal the actual influence of the urban road sound environment on human emotions and cognition.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of environmental assessment, and in particular to a road sound environment emotional influence evaluation method and device. BACKGROUND

[0002] With the continuous acceleration of urbanization, various sound sources such as traffic flow, commercial activities and construction noise in urban road areas are intertwined and superimposed, forming a complex structure and time-space dynamic sound environment. This multi-dimensional sound field not only puts forward higher requirements for objective indicators such as frequency spectrum distribution and equivalent sound pressure level in the physical quantity measurement level, but also has a significant impact on the emotional experience and cognitive state of residents. However, existing road noise monitoring and evaluation mainly focuses on the measurement and statistical analysis of acoustic parameters, which is difficult to accurately capture the subjective feelings of individuals in actual scenarios such as comfort and annoyance, and therefore, it is necessary to more comprehensively and accurately reveal the actual influence of urban road sound environment on human emotions and cognition. SUMMARY

[0003] The present application provides a road sound environment emotional influence evaluation method and device, which solves the technical problem that existing road noise monitoring and evaluation mainly focuses on the measurement and statistical analysis of acoustic parameters, and is difficult to comprehensively and accurately reveal the actual influence of urban road sound environment on human emotions and cognition.

[0004] The present application provides a road sound environment emotional influence evaluation method and device, which solves the technical problem that existing road noise monitoring and evaluation mainly focuses on the measurement and statistical analysis of acoustic parameters, and is difficult to comprehensively and accurately reveal the actual influence of urban road sound environment on human emotions and cognition.

[0005] In the urban road sound environment, the effective dominant audio and the dominant sound pressure level of a plurality of road dominant sound source types are determined, and the road dominant sound source types are classified into a plurality of road noise types according to the sound source attributes;

[0006] The effective dominant audio of each road dominant sound source type is embedded to build a questionnaire platform, and the subjective scores of a plurality of subjective indicators are collected through the questionnaire platform;

[0007] The brain electrical signals are collected by triggering the associated effective dominant audio according to each road noise type;

[0008] The subjective key principal components of each road noise type are determined based on the subjective scores associated with each road noise type respectively through principal component analysis;

[0009] A plurality of spectral features of each brain electrical signal are extracted, and the spectral features are analyzed by principal component analysis according to the road noise types, and the objective key principal components of each road noise type are determined;

[0010] Respectively, the subjective key principal component, the objective key principal component, the effective dominant audio and the dominant sound pressure level are used for comprehensive evaluation to determine the emotion comprehensive index of each road noise type.

[0011] Optionally, the sampling in the urban road sound environment, determining the effective dominant audio and the dominant sound pressure level of the plurality of road dominant sound source types, comprises:

[0012] In the road area of one or more types of urban road sound environment functional areas, sampling points are arranged to synchronously collect audio samples and time sequence sound pressure levels;

[0013] The audio samples are subjected to sound event boundary cutting to determine corresponding sub-audios;

[0014] According to the dominant frequency values of the sub-audios, the plurality of road dominant sound source types are classified according to preset dominant sound source types;

[0015] The dominant frequency confidence intervals of each road dominant sound source type are calculated based on the dominant frequency values of each road dominant sound source type, and the effective dominant frequency audios of each road dominant sound source type are screened out according to the dominant frequency confidence intervals.

[0016] The dominant sound pressure levels of the effective dominant frequency audios are determined based on the time sequence sound pressure levels.

[0017] Optionally, the audio samples are subjected to sound event boundary cutting to determine corresponding sub-audios, comprising:

[0018] The audio samples are subjected to audio preprocessing by Adobe Audition software, and initialized audios are correspondingly outputted;

[0019] According to the Melville spectrogram analysis of each initialized audio, a plurality of sub-audios are cut out.

[0020] Optionally, the effective dominant audios of each road dominant sound source type are embedded to build a questionnaire platform, comprising:

[0021] An interactive front-end page of the questionnaire platform is built based on Html, CSS and JavaScript; wherein, the page framework is built by Html, the questionnaire form is designed in combination with the subjective indicators, the effective dominant audios of each road dominant sound source type are imported based on the Audio component, and the audio scene pictures associated with the audios are imported based on the image component, the page style is designed by CSS, and the page interaction is designed by JavaScript in combination with the Audio component ID.

[0022] The interactive front-end page and the MySQL database are deployed through the cloud server, a subjective index standardization form is established in the MySQL database, Java is used to connect the interactive front-end page and the MySQL database, and the questionnaire platform is determined.

[0023] Optionally, the extracting the multiple spectrum features of each of the brain electrical signals comprises:

[0024] The brain electrical signals are filtered and artifact removed, and clean brain electrical signals are outputted.

[0025] According to the power spectrum density of each of the clean brain electrical signals, the power in different frequency bands is determined as spectrum features; the frequency bands comprise Theta waves, Alpha waves and Beta waves.

[0026] Optionally, the respectively performing comprehensive evaluation on the associated subjective key principal component, the objective key principal component, the effective dominant audio and the dominant sound pressure level to determine the emotion comprehensive index of each of the road noise types comprises:

[0027] According to the road noise type, the dominant frequency value of the associated effective dominant audio is subjected to average value operation and logarithm is taken to determine the corresponding logarithmic dominant frequency value.

[0028] According to the road noise type, the associated dominant sound pressure level is subjected to average value operation to output the corresponding sound pressure level average value.

[0029] The subjective key principal component and the objective key principal component are respectively subjected to average value operation to determine the subjective key principal component average value and the objective key principal component average value.

[0030] The subjective key principal component average value, the objective key principal component average value, the logarithmic dominant frequency value and the sound pressure level average value are subjected to weighted operation to determine the emotion comprehensive index of each of the road noise types.

[0031] The second aspect of the present application provides a road sound environment emotion influence evaluation device, comprising:

[0032] The data sampling module is used for sampling in the urban road sound environment, determining the effective dominant audio and the dominant sound pressure level of multiple road dominant sound source types, and classifying each of the road dominant sound source types into multiple road noise types according to sound source attributes.

[0033] The subjective evaluation module is used for embedding the effective dominant audio of each of the road dominant sound source types into a questionnaire platform, and collecting subjective scores of multiple subjective indexes through the questionnaire platform.

[0034] The objective evaluation module is used for triggering the collection of brain electrical signals according to the associated effective dominant audio of each of the road noise types.

[0035] a subjective analysis module configured to perform principal component analysis on the subjective scores associated with each of the road noise types, respectively, to determine subjective key principal components corresponding to each of the road noise types;

[0036] an objective analysis module configured to extract a plurality of spectral features of each of the electroencephalogram signals, perform principal component analysis on the spectral features according to the road noise types, and determine objective key principal components corresponding to each of the road noise types;

[0037] an emotion evaluation module configured to perform comprehensive evaluation on the subjective key principal components, the objective key principal components, the effective dominant audio and the dominant sound pressure level associated with each of the road noise types, respectively, to determine an emotion comprehensive index of each of the road noise types.

[0038] The third aspect of the present application provides a computer device, comprising a memory and a processor, the memory stores a computer program, and the computer program is executed by the processor to make the processor perform the steps of the road sound environment emotion influence evaluation method according to any one of the above aspects.

[0039] The fourth aspect of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed to implement the road sound environment emotion influence evaluation method according to any one of the above aspects.

[0040] The fifth aspect of the present application provides a computer program product, which comprises computer programs / instructions, and the computer programs / instructions are executed by a processor to implement the road sound environment emotion influence evaluation method according to any one of the above aspects.

[0041] From the above technical solutions, the present application has the following advantages:

[0042] The above scheme of the present application provides a road sound environment emotion influence evaluation method, comprising: sampling in an urban road sound environment, determining effective dominant audio and dominant sound pressure level of a plurality of road dominant sound source types, and classifying each road dominant sound source type into a plurality of road noise types according to the sound source attribute; embedding the effective dominant audio of each road dominant sound source type to build a questionnaire platform, and collecting subjective scores of a plurality of subjective indicators through the questionnaire platform; collecting electroencephalogram signals according to the associated effective dominant audio of each road noise type; performing principal component analysis based on the subjective scores associated with each road noise type, respectively, to determine the subjective key principal components of each road noise type; extracting a plurality of spectral features of each electroencephalogram signal, performing principal component analysis on the spectral features according to the road noise type, and determining the objective key principal components of each road noise type; and performing comprehensive evaluation by using the associated subjective key principal components, objective key principal components, effective dominant audio and dominant sound pressure level, respectively, to determine the emotional comprehensive index of each road noise type. Based on the above scheme, a large amount of typical data is collected in the urban road sound environment, the questionnaire platform is built by embedding the effective dominant audio to guide the user to perform audio subjective perception scoring, and the objective emotion indicators are quantified by using the electroencephalogram signal response of the sound source stimulation. Overall, by fusing environmental psychology and neuroscience means, taking into account subjective and objective data and sound source information, a comprehensive evaluation system of the emotional suitability of the urban road sound environment is formed, which can effectively distinguish the influence of different road sound sources on the emotional fluctuation of the crowd, so as to more comprehensively and accurately reveal the actual influence of the urban road sound environment on human emotions and cognition. BRIEF DESCRIPTION OF DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0044] Figure 1 A step flow chart of a road sound environment emotion influence evaluation method provided by the embodiment of the present application;

[0045] Figure 2 A city road sound environment data sampling flow chart provided by the embodiment of the present application;

[0046] Figure 3 A questionnaire platform design schematic diagram provided by the embodiment of the present application;

[0047] Figure 4 A subjective evaluation flow chart provided by the embodiment of the present application;

[0048] Figure 5A brain electrical index objective evaluation flowchart provided by the embodiment of the present application is shown in the figure.

[0049] Figure 6 An emotion comprehensive index calculation flowchart provided by the embodiment of the present application is shown in the figure.

[0050] Figure 7 A structural block diagram of a road sound environment emotion influence evaluation device provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0051] The embodiment of the present application provides a road sound environment emotion influence evaluation method and device, and aims to solve the technical problem that the existing road noise monitoring and evaluation are mostly focused on the measurement and statistical analysis of acoustic parameters, and it is difficult to comprehensively and accurately reveal the actual influence of the urban road sound environment on human emotions and cognition.

[0052] In order to make the purposes, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely in combination with the accompanying drawings. Obviously, the following described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0053] Please refer to Figure 1 , Figure 1 A step flowchart of a road sound environment emotion influence evaluation method provided by the embodiment of the present application is shown in the figure.

[0054] The road sound environment emotion influence evaluation method provided by the embodiment includes:

[0055] Step 101, sampling in the urban road sound environment, determining the effective dominant audio and the dominant sound pressure level of a plurality of road dominant sound source types, and classifying each road dominant sound source type into a plurality of road noise types according to the sound source attribute.

[0056] The road dominant sound source type refers to the sound source category of a sound event occurring in the urban road sound environment. The sound event refers to a sound signal with a certain acoustic scene generated by a specific sound source, usually with a clear start and end time.

[0057] The effective dominant audio refers to an audio segment that can effectively represent a certain type of sound event in the urban road sound environment.

[0058] The dominant sound pressure level refers to the sound pressure level feature of the effective dominant audio.

[0059] Sound source attribute refers to a certain type of road sound source characteristic described based on a sound-emitting subject or acoustic characteristics.

[0060] Road noise type refers to a noise type obtained by classifying audio clips of the urban road sound environment according to sound source attributes.

[0061] It should be noted that a plurality of sampling points are arranged in the road environment of the city, audio is collected at each sampling point by a recording pen or the like, sound pressure level is collected by a high-precision A-weighted sound level meter or the like, and then effective dominant audio and dominant sound pressure level belonging to a plurality of road dominant sound source types are determined according to the collected data, and each road dominant sound source type is classified according to sound source attributes to determine a plurality of road noise types.

[0062] In one specific embodiment of the present embodiment, sampling is performed in the urban road sound environment, and effective dominant audio and dominant sound pressure level belonging to a plurality of road dominant sound source types are determined, including:

[0063] In the road area of one or more types of urban road sound environment functional areas, sampling points are arranged to synchronously collect audio samples and time-series sound pressure levels;

[0064] The sound event boundary of each audio sample is cut to determine the corresponding sub-audio;

[0065] According to the dominant frequency value of each sub-audio, the preset dominant sound source type is classified to determine a plurality of road dominant sound source types;

[0066] The dominant frequency confidence interval is calculated based on the dominant frequency value belonging to each road dominant sound source type, and the effective dominant frequency audio of each road dominant sound source type is filtered according to the dominant frequency confidence interval.

[0067] The dominant sound pressure level of the corresponding effective dominant frequency audio is determined based on each time-series sound pressure level.

[0068] Urban road sound environment functional area refers to a sound environment functional area obtained by dividing the road environment of the city according to the national sound environment functional area zoning standard.

[0069] Audio sample refers to an original audio file collected at a sampling point.

[0070] Time-series sound pressure level refers to a sequence composed of sound pressure levels recorded over time.

[0071] Sound event boundary refers to a time duration period composed of the start and end times of a sound event.

[0072] Dominant sound source type refers to the sound source category of different sound events.

[0073] The dominant frequency value refers to the highest frequency peak within an audio time period.

[0074] The dominant frequency confidence interval refers to the range of dominant frequency values ​​used to select stable and representative audio segments.

[0075] It should be noted that, as Figure 2 As shown, this embodiment identifies six categories of urban road noise environment functional zones based on the national noise environment functional zoning classification: Category 0 refers to areas that require special quietness, and this embodiment selects natural scenic areas; Category 1 refers to areas that require quietness, such as residential areas, medical and health facilities, and cultural and educational facilities, and this embodiment identifies residential areas and teaching areas; Category 2 refers to areas that require quietness, such as financial and trade areas and market areas, and this embodiment identifies commercial areas; Category 3 refers to areas that require quietness, such as industrial production, warehousing and logistics areas, and this embodiment identifies construction areas; Category 4 refers to areas that require quietness, such as main traffic arteries. In this embodiment, the area where traffic noise has a serious impact on the surrounding environment is identified as a transportation hub area. Multiple sampling points are deployed in the above six urban road acoustic environment functional zones to ensure sufficient environmental sound pressure level samples, while the audio collection also ensures the richness of the dominant sound source types. In specific implementation, sampling points can be set up in the road areas of one or more types of urban road acoustic environment functional zones. Each sampling point in the urban road acoustic environment functional zone can be set up 100 meters apart. Data collection is carried out during the daytime period on weekdays. The recording time for each sampling point is 5 minutes, the sampling interval for sound pressure level is 1 second, and the audio sampling frequency is 32000Hz.

[0076] The collected audio samples are subjected to various sound event recognition to determine the duration of each sound event and extract the corresponding sub-audio. A Fast Fourier Transform (FFT) is performed on the sub-audio, and the dominant frequency value of each sub-audio is extracted using the scipy and numpy libraries in Python. The sub-audio are classified according to the dominant sound source type and its corresponding standard frequency range based on the dominant frequency value, identifying multiple dominant sound source types for each road, while ensuring that each dominant sound source type has multiple samples. For each dominant sound source type, the dominant frequency values ​​of the sub-audio are grouped into a sample dominant frequency range, and a confidence interval (e.g., 95% confidence interval) is calculated for the dominant frequency value. Representative sub-audio, i.e., effective dominant frequency audio, is then extracted from the corresponding sample dominant frequency range according to the dominant frequency confidence interval. The effective dominant frequency audio can be numbered according to different dominant sound source types for subsequent subjective and objective evaluation collection.

[0077] The effective main frequency audio can extract the corresponding sound pressure level data from the corresponding time sequence sound pressure level, and the corresponding sound pressure level feature is obtained by processing, that is, the dominant sound pressure level, which can include: , , , , or , represent the maximum sound pressure level of a piece of audio, represent the minimum sound pressure level of a piece of audio, and , , respectively represent the average peak value, average value and background value of the noise, represent the average non-steady-state sound energy in a period of time, which is the continuous equivalent A sound level, which can be obtained by formula (1):

[0078] (1)

[0079] In the formula, represent the number of sound pressure levels, represent the first sound pressure level; in this embodiment may be 300, representing the sound pressure level in 5 minutes, and the interval time is 1 second.

[0080] Exemplarily, the road dominant sound source types include 15 types of bird chirping sound, wind sound, rain sound, dog barking sound, human speaking sound, whistle sound, bus sound, large vehicle sound, broadcast sound, motorcycle sound, construction sound, car sound, bicycle sound, airplane sound and high-speed rail sound. The road dominant sound source types are classified and determined as road noise types, which can be: bird chirping sound, wind sound, rain sound and dog barking sound are classified as natural noise, human speaking sound, whistle sound and broadcast sound are classified as social noise, bus sound, large vehicle sound, motorcycle sound, car sound, bicycle sound, airplane sound and high-speed rail sound are classified as traffic noise, and construction sound is classified as building noise.

[0081] In a more specific embodiment of the present embodiment, the sound event boundary of each audio sample is cut to determine the corresponding sub-audio, which includes:

[0082] Each audio sample is preprocessed by Adobe Audition software, and the initialized audio is output correspondingly;

[0083] According to the Melville spectrum diagram analysis of each initialized audio, a plurality of sub-audios are cut.

[0084] Adobe Audition software is a professional audio processing tool supporting audio editing, noise reduction, spectrum analysis and other functions, which can be referred to in the prior art.

[0085] The Melville spectrum is a two-dimensional image that maps an audio signal onto a Mel scale, with time on the horizontal axis, Mel frequency on the vertical axis, and color intensity representing energy.

[0086] It should be noted that, in this embodiment, when determining the sub-audio, the collected audio samples are first visually monitored using Adobe Audition software to eliminate occasional interference during the collection, capture noise samples, and adjust the noise reduction percentage to make the audio quality clear and recognizable. After obtaining the initial audio, the Melville spectrogram feature analysis is performed on each initial audio. By analyzing the energy level, the time-frequency characteristics of different sound sources can be clearly distinguished, thereby identifying various sound events. Then, the audio is segmented to obtain multiple sub-audio.

[0087] Step 102: Build a questionnaire platform by embedding effective dominant audio of each road's dominant sound source type, and collect subjective scores of multiple subjective indicators through the questionnaire platform.

[0088] A questionnaire platform refers to an interactive system that collects users' subjective feedback through questionnaires.

[0089] Subjective metrics refer to the metric dimensions that quantify users' subjective feelings about the sound environment.

[0090] Subjective rating refers to the quantitative evaluation results given by users on subjective indicators based on their own feelings, which can be expressed as specific numerical values ​​or grade evaluations.

[0091] It should be noted that this embodiment establishes a questionnaire platform. By embedding effective dominant audio of each road's dominant sound source, it guides users to score according to subjective indicators based on their subjective perception of the audio of each road's dominant sound source type, thereby achieving standardized collection of subjective scoring data. In specific implementation, the designed subjective indicators may include sound comfort, sound pleasure, sound loudness, sound annoyance, and sound emotional perception. The scoring can be quantified using a five-point Likert scale.

[0092] In one specific embodiment of this example, a questionnaire platform is built using effective dominant audio embeddings of each road's dominant sound source type, including:

[0093] The interactive front-end page of the questionnaire platform is built using HTML, CSS, and JavaScript. Specifically, the page framework is constructed using HTML and the questionnaire form is designed in combination with subjective indicators. Effective dominant audio of each road's dominant sound source type is imported based on the Audio component, and related audio scene images are imported based on the image component. The page style is designed using CSS, and the page interaction is designed in combination with the Audio component ID using JavaScript.

[0094] The subjective index standardization form is a structured table used to store and manage subjective scores of various subjective indexes, ensuring the standardization of data format, field meaning, and mapping relationship.

[0095] The subjective index standardization form is a structured table used to store and manage subjective scores of various subjective indexes, ensuring the standardization of data format, field meaning, and mapping relationship.

[0096] It should be noted that, as shown in Figure 3 The process of building a questionnaire platform using a Web platform is divided into front-end and back-end:

[0097] First, use VS Code to write Html, Css, and Javascript codes to deploy the interactive front-end page. The interactive front-end page includes the main interface and the questionnaire interface. The main interface is used to introduce website information and questionnaire process to users, and the questionnaire interface includes questionnaire forms designed in combination with subjective indexes, which are used for users to perform subjective scoring after listening to audio. Html is the framework of the page, responsible for the design and layout of various components. Audio components can be used to import effective dominant audio of various road dominant sound source types, and image components can be used to import audio scene pictures associated with effective dominant audio to introduce road dominant sound sources. Css is the style of the page, responsible for font, color, dynamic effect, and interface responsive design, etc. For each component of Html, CSS can define, for example, font, color, etc., so that a simple and beautiful page can be designed. Javascript is responsible for page interaction, such as uploading forms to the back-end database, and assigning independent IDs to button components of Audio components for each road dominant sound source type to realize questionnaire page jump (such as assigning 1 to bird chirping, and so on) and performing page jump through connection URL. It can be understood that the jump between the main interface and the questionnaire interface is based on the button component of Html, and the url assigned by JavaScript code is used to distinguish the questionnaire interface of different audio, so as to distinguish the questionnaire filling data of different dominant sound source types by users, reducing the repetitiveness of front-end code. At the same time, the interactive page can also be designed responsively, which can match the size of different device ends, such as mobile phone end and PC end.

[0098] The core of the back-end design is to collect and analyze user data through a suitable cloud server (such as Alibaba Cloud ECS) and a database Mysql. Log in to the cloud server to download the required libraries such as nginx and php, use the cloud server to deploy the front-end designed code, and access it through the public IP. Then set up a database, build a standardized form for subjective indicators, open port 3306, and you can access and modify the remote database. Finally, connect the front-end deployment through Java to upload the user-filled data for collection and storage.

[0099] It is worth mentioning that using the Web platform to build a questionnaire platform to publish questionnaires can spread the questionnaire more quickly. Users can log in using only mobile devices. For user groups, the convenience of accessing the questionnaire platform ensures the universality of the user group. For the filling of personal information, the questionnaire will include age, gender, and living environment. This ensures the diversity of the questionnaire sample. When playing audio such as "bird chirping," "rain sound," and "construction sound," high-definition pictures of the corresponding scene will be displayed to help users "immerse" in the real scene and reduce the abstract perception bias when subjective scoring. After listening to the audio, the user will be guided to fill out the questionnaire. The forms of each subjective indicator will be uploaded to the cloud server's database Mysql through the back-end code. The database provides formatted subjective indicator standardized forms to facilitate data visualization on different software. In this way, for different types of road dominant sound sources, there will be a separate questionnaire rating sample big data, as shown in Figure 4 After obtaining the data, first perform data cleaning and preprocessing, detect whether there is missing data in the back-end database Mysql, and if there is missing data, fill in the data table with the mean value. Perform reliability testing on the Likert scale and calculate the Cronbach's alpha coefficient, which reflects the consistency of each item in the scale in measuring the same latent variable. When the number of valid questionnaires corresponding to the road dominant sound source type is ≥30, the consistency of each indicator can be estimated more robustly. The expression of Cronbach's alpha coefficient is as follows:

[0100] (2)

[0101] where, is the number of scale items (here the number of subjective indicators corresponds to the number of subjective indicators), is the variance of the th item, is the variance of the total score of all items. When , it indicates that the reliability of the questionnaire indicator is acceptable. At this time, it is indicated that the questionnaire index has good reliability; meanwhile, the psychological index data obtained by the Likert scale can be subjected to descriptive statistics, and the mean, median and the like of the index under various sound sources are calculated to summarize the overall reaction trend of the emotion under different sound sources.

[0102] It can be understood that, in order to facilitate subsequent statistics in units of road noise types, a questionnaire form can be designed in combination with the subjective index in units of road noise types, and the effective dominant audio of each road dominant sound source type can be imported in units of road noise types based on the Audio component.

[0103] In step 103, the effective dominant audio associated with each road noise type is used to trigger the collection of the electroencephalogram signal.

[0104] It should be noted that, in this embodiment, the effective dominant audio of each road noise type (such as traffic noise, social noise, natural noise and building noise) is played, and a high-precision electroencephalograph and a high-quality headset are used to measure the electroencephalogram (EEG) signal of the user.

[0105] In specific implementation, the number of required experiments is confirmed by using the T test of Gpower, the experimental groups are balanced designed by using the Latin square, and the number of people in each group is set according to the type and number of audio, for example, 4 people are set as a group to correspond to 4 road noise types, the software used in the measurement process of the electroencephalogram signal can be Emotiv and E-prime, the playing time point of each audio is controlled by E-prime script writing, the electroencephalogram signal in the same time period is recorded and distinguished, the participants need to wear the EEG headset (such as high-precision electroencephalograph and high-quality headset) to ensure good contact between the electrode and the scalp to reduce the instability of the signal, the electroencephalogram signal is recorded by using the Emotiv Pro software after the participants are exposed to the noise stimulus of the audio, and the effective dominant audio with a 2-minute exposure time is preferably ensured; before each stimulation, the participants will have a short adaptation period to ensure that they are in a relaxed state, there will be a two-minute resting period in the stimulation interval, during the experiment, all participants are required to remain quiet to reduce the interference of artifacts, in order to ensure the quality of the data, the EEG signal of each participant will be standardized during the experiment, and the electrode contact quality is verified, after each sound stimulation, the EEG signal will be automatically stored, and a quick check is performed at the end of each stimulation cycle to ensure that the data is complete and accurate.

[0106] It can be understood that white noise as a control group can also be set, that is, the effective dominant audio associated with each road noise type is used to trigger the collection of electroencephalogram signals, and the white noise helps to distinguish whether the difference in electroencephalogram signals under different road noise stimulation is only derived from the effective dominant audio and not irrelevant acoustic interference. The white noise can be spatial stereo sound generated by Adobe Audition, with a sampling rate of 32000hz and a bit depth of 16bit, consistent with the aforementioned recording device collection conditions.

[0107] Step 104, respectively based on the subjective scores associated with each road noise type, principal component analysis is performed to determine the subjective key principal components of each road noise type.

[0108] Subjective key principal components refer to principal components in the subjective dimension that can explain most of the variability of the data.

[0109] It should be noted that the classification of each road dominant sound source type into the corresponding road noise type can first use ANOVA to compare the effects of different road noise types on sound comfort, sound pleasure and other dimensions, and test whether there are significant differences between different road noise types.

[0110] After data analysis, as shown in Figure 4 , the mean and standard deviation of the score of each road noise type and each subjective index dimension are calculated to understand the average perception of each noise and the volatility of the evaluation, and then principal component analysis (PCA) is performed to reduce the dimension and find the dominant evaluation dimension: 1, the questionnaire scores of each subjective index are standardized; 2, the principal component analysis of each subjective index dimension is performed to identify several principal components that best represent the overall comfort change of the user; 3, combined with the variance contribution rate of each principal component, the subjective key principal component of the subjective evaluation is determined.

[0111] Step 105, extract multiple spectral features of each electroencephalogram signal, and perform principal component analysis on the spectral features according to the road noise type to determine the objective key principal component of each road noise type.

[0112] Spectral features refer to quantitative indicators that can represent the state of brain electrical activity extracted from electroencephalogram signals through frequency domain analysis.

[0113] Objective key principal components refer to principal components in the objective dimension that can explain most of the variability of the data.

[0114] It should be noted that multiple spectral features are extracted from electroencephalogram signals through frequency domain analysis, and principal component analysis is performed on the associated spectral features in units of road noise type to determine the objective key principal component.

[0115] ​In one specific embodiment of the present embodiment, a plurality of spectral features of each brain electrical signal are extracted, including:

[0116] Each brain electrical signal is filtered and artifact removed to output a clean brain electrical signal.

[0117] According to the power spectrum density of each clean brain electrical signal, the power in different frequency bands is determined as the spectral feature; the frequency bands include Theta wave, Alpha wave and Beta wave.

[0118] It should be noted that, as Figure 5 shown, after the EEG data is collected, the Mne library in Python software can be used for signal preprocessing first, such as using a filter to remove noise in the brain electrical signal, the range is 1-30 hz, then artifact analysis is performed on the brain electrical signal, independent component analysis (ICA) is used to remove electrooculogram, electromyogram and other artifacts; then the clean brain electrical signal obtained by processing is subjected to feature extraction, short-time Fourier transform is used to analyze the brain electrical signal under different road noise types, the frequency spectrum under each sound stimulus is obtained, the influence of different sound sources on the frequency component of the brain electrical wave is observed, and the power spectrum density of each brain electrical signal is calculated, finally, according to the frequency bands of Theta wave (4-8 Hz), Alpha wave (8-3 Hz) and Beta wave (13-30 Hz), the power (such as alpha wave absolute power, beta wave relative power, theta wave relative power) is calculated based on the corresponding power spectrum density as the spectral feature, which can analyze which wave band changes most significantly under different sound sources.

[0119] It can be understood that the present embodiment introduces EEG brain electrical experiment as an objective evaluation system, because compared with traditional questionnaire survey or subjective evaluation, brain electrical data can directly reflect the instantaneous reaction of human brain to sound environment, avoiding subjective bias, therefore, the objective system of the present embodiment quantitatively evaluates the influence of noise on cognitive load, emotional stress and attention through specific indicators (power) of brain electrical signal, making the evaluation more objective; specifically, different frequencies, intensities and types (such as natural noise, traffic noise) of sound can be directly related to the changes of brain electrical response, helping to deeply understand the influence of sound characteristics on brain, so as to obtain an objective score.

[0120] Step 106, respectively using the associated subjective key principal component, objective key principal component, effective dominant audio and dominant sound pressure level for comprehensive evaluation to determine the emotional comprehensive index of each road noise type.

[0121] The emotional comprehensive index refers to a quantitative index representing the emotional perception fluctuation state by comprehensive analysis of sound source information and subjective and objective index data.

[0122] It can be understood that in step 101, if single type of urban road sound environment function area is sampled and evaluated separately, the emotion comprehensive index can evaluate the emotional influence of different road noise types of noise sources on users in a certain type of urban road sound environment function area, and if multiple types of urban road sound environment function areas are sampled and evaluated, the emotion comprehensive index can evaluate the emotional influence of different road noise types of noise sources on users in the urban road sound environment as a whole, so the present scheme can evaluate the sound environment emotional influence at the local level of the function area and at the global level of the city. The higher the value of the emotion comprehensive index, the greater the intensity of the emotional fluctuation. For example, if the natural noise score is 8.73, the traffic noise score is 10.29, the social noise score is 9.91, and the building noise score is 10.20, it indicates that in the road environment, the influence degree of natural noise on emotional fluctuation is the smallest, the influence degree of traffic noise and building noise on human emotional fluctuation is the largest, and the influence degree of social noise on emotion is an intermediate value.

[0123] In one specific embodiment of the present embodiment, step 106 includes the following sub-steps:

[0124] The dominant frequency value of the effective dominant audio of each road noise type is averaged and logarithmized to determine the corresponding logarithmic dominant frequency value;

[0125] The associated dominant sound pressure level is averaged according to each road noise type, and the corresponding sound pressure level average is output;

[0126] The subjective and objective key principal components are averaged respectively to determine the subjective and objective key principal component averages;

[0127] The subjective and objective key principal component averages, the logarithmic dominant frequency value, and the sound pressure level average are weighted to determine the emotion comprehensive index of each road noise type.

[0128] It should be noted that according to the sound-brain-emotion model, the dominant sound pressure level, the subjective and objective key principal components, and the dominant frequency value are averaged and logarithmized, and then the weighted sum of the above operation results is obtained to obtain the corresponding emotion comprehensive index.

[0129] In a more specific embodiment of the present embodiment, in combination with Figure 6 As shown in the figure, the calculation process of the emotion comprehensive index includes:

[0130] (3)

[0131] wherein, wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows: wherein k represents the kth road noise type (k = 1, 2, 3, 4, determined according to the number of road noise types), and wherein the road noise types are as follows:

[0132] Further, the present embodiment calculates the weights according to the variance of the principal components, wherein the first subjective key principal component accounts for and the second subjective key principal component accounts for in the subjective experiment, and wherein the first objective key principal component accounts for and the second objective key principal component accounts for in the objective experiment, and wherein the weight calculation formula is:

[0133] (4)

[0134] (5)

[0135] As for the frequency weight and the sound pressure level weight, first, the frequency and the sound pressure level of the effective dominant audio are normalized according to the road noise type, and the normalized frequency and the normalized sound pressure level are determined, then the corresponding frequency weight and the sound pressure level weight ; wherein, the calculation process of Min-Max normalization for the dominant sound pressure level and the dominant frequency value of the effective dominant audio eliminates the difference of different dimensions:

[0136] (6)

[0137] wherein, is the standardized value, the interval is [0-1]; is the initial value of the data; is the maximum value in the list data; is the minimum value in the list data.

[0138] In the embodiment of the present application, firstly, the sampling point is set in the urban road sound environment to synchronously collect the time sequence audio and sound pressure level data, a large number of typical dominant audios can be collected in the road environment, and the sufficiency of the data sample guarantees the accuracy and effectiveness of the comprehensive evaluation system of the method;Secondly, based on the dominant frequency confidence interval, the effectiveness of the sound source is screened, the screened multiple road dominant sound source types are classified into different road noise types, and the subjective perception questionnaire is issued through the questionnaire platform to obtain the subjective emotion evaluation under different sound environments, the online questionnaire is used to replace the traditional offline "field" questionnaire, the form data filled by the user can be efficiently and conveniently collected to realize the analysis of the subjective evaluation system under the big data, and meanwhile, there are independent questionnaire rating sample big data for the dominant sound source types, which is helpful to realize the difference of the indicators of different sound sources and improve the accuracy of the subjective evaluation system;Meanwhile, under the controlled experimental conditions, the electroencephalograph is used to record the electroencephalogram response of each sound source group to stimulate the quantitative objective emotion index, which avoids the subjective deviation of the questionnaire and quantifies the influence of noise on cognitive load, emotional stress and attention;Finally, the comprehensive emotion index is obtained by comprehensively analyzing the road noise type, sound source information and subjective and objective index data, the quality of the sound environment is exhibited from multiple angles, and the one-sidedness of single dimension is avoided;Therefore, as a whole, by fusing the environmental psychology and neuroscientific means, the subjective and objective data and sound source information are considered, the comprehensive evaluation system of the urban road sound environment emotion suitability is formed, the influence of different road sound sources on the emotional fluctuation of the crowd can be effectively distinguished, the actual influence of the urban road sound environment on the human emotion and cognition is more comprehensively and accurately revealed, and reliable quantitative decision support is provided for the formulation of noise control strategy and the optimization of urban acoustic environment.

[0139] Please refer to Figure 7 , Figure 7 is a structural block diagram of a road sound environment emotion influence evaluation device provided by the embodiment of the present application.

[0140] The road sound environment emotion influence evaluation device provided by the present application comprises:

[0141] The data sampling module 701 is configured to sample in the urban road sound environment, determine effective dominant audio and dominant sound pressure level of a plurality of road dominant sound source types, and classify each road dominant sound source type into a plurality of road noise types according to sound source attributes;

[0142] The subjective evaluation module 702 is configured to build a questionnaire platform by embedding the effective dominant audio of each road dominant sound source type, and collect subjective scores of a plurality of subjective indicators through the questionnaire platform;

[0143] The objective evaluation module 703 is configured to trigger the collection of electroencephalogram signals according to the associated effective dominant audio of each road noise type;

[0144] The subjective analysis module 704 is configured to perform principal component analysis based on the subjective scores associated with each road noise type, and determine subjective key principal components of each road noise type;

[0145] The objective analysis module 705 is configured to extract a plurality of spectral features of each electroencephalogram signal, perform principal component analysis on the spectral features according to the road noise type, and determine objective key principal components of each road noise type;

[0146] The emotion evaluation module 706 is configured to perform comprehensive evaluation by using the associated subjective key principal components, objective key principal components, effective dominant audio and dominant sound pressure level, and determine an emotion comprehensive index of each road noise type.

[0147] Further, in the urban road sound environment, the effective dominant audio and the dominant sound pressure level of a plurality of road dominant sound source types are determined, including:

[0148] In one or more functional areas of the urban road sound environment, a sampling point is set to synchronously collect audio samples and time-series sound pressure levels;

[0149] The sound event boundary of each audio sample is cut to determine the corresponding sub-audio;

[0150] According to the dominant frequency value of each sub-audio, the plurality of road dominant sound source types are classified according to the preset dominant sound source type;

[0151] The dominant frequency confidence interval of each road dominant sound source type is calculated based on the dominant frequency value of each road dominant sound source type, and the effective dominant frequency audio of each road dominant sound source type is selected according to the dominant frequency confidence interval;

[0152] The dominant sound pressure level of the effective dominant frequency audio is determined based on the time-series sound pressure level.

[0153] Further, the sound event boundary of each audio sample is cut to determine the corresponding sub-audio, including:

[0154] The audio preprocessing is performed on each audio sample by using the Adobe Audition software, and the initialized audio is output correspondingly;

[0155] According to the Melville spectrum analysis of each initialized audio, a plurality of sub-audios are cut out.

[0156] Further, the effective dominant audio embedding of each road dominant sound source type is used to build a questionnaire platform, including:

[0157] The interactive front-end page of the questionnaire platform is built based on Html, CSS and JavaScript; wherein, the page framework is built by Html, and the questionnaire form is designed in combination with the subjective indicators, the effective dominant audio of each road dominant sound source type is imported based on the Audio component, and the associated audio scene picture is imported based on the image component, the page style is designed by CSS, and the Audio component ID is assigned by JavaScript and the page interaction is designed in combination with the Audio component ID;

[0158] The interactive front-end page and the MySQL database are deployed through the cloud server, the subjective indicator standardized form is established in the MySQL database, the interactive front-end page and the MySQL database are connected by using Java, and the questionnaire platform is determined.

[0159] Further, a plurality of spectral features of each electroencephalogram signal are extracted, including:

[0160] The electroencephalogram signal is filtered and artifact removed, and a clean electroencephalogram signal is output;

[0161] According to the power spectral density of each clean electroencephalogram signal, the power in different frequency bands is determined as the spectral feature; the frequency bands include Theta wave, Alpha wave and Beta wave.

[0162] Further, the emotion evaluation module 706 is specifically used for:

[0163] The main frequency value of the effective dominant audio is averaged according to each road noise type, and the logarithm of the main frequency value is determined;

[0164] The associated dominant sound pressure level is averaged according to each road noise type, and the sound pressure level average value is output;

[0165] The subjective key principal components and the objective key principal components are averaged respectively, and the subjective key principal component average value and the objective key principal component average value are determined;

[0166] The subjective key principal component average value, the objective key principal component average value, the logarithm main frequency value and the sound pressure level average value are weighted, and the emotion comprehensive index of each road noise type is determined.

[0167] The embodiment of the present application further provides a computer device, comprising a memory and a processor, the memory storing a computer program; the computer program is executed by the processor, so that the processor executes the steps of the road sound environment emotion influence evaluation method according to any one of the above embodiments.

[0168] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program / instruction, and the computer program / instruction is executed by a processor to realize the steps of the road sound environment emotion influence evaluation method according to any one of the above embodiments.

[0169] The embodiment of the present application further provides a computer program product, comprising a computer program / instruction, and the computer program / instruction is executed by a processor to realize the steps of the road sound environment emotion influence evaluation method according to any one of the above embodiments.

[0170] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and module can refer to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0171] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms.

[0172] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. According to actual needs, some or all of the units can be selected to achieve the purpose of the embodiment scheme.

[0173] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0174] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the entire or part of the technical solutions that essentially contribute to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0175] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A road sound environment emotion influence evaluation method characterized by comprising: The method comprises the following steps: sampling in an urban road sound environment to determine effective dominant audio and dominant sound pressure levels of a plurality of road dominant sound source types, and classifying each of the road dominant sound source types into a plurality of road noise types according to sound source attributes; embedding the effective dominant audio of each of the road dominant sound source types into a questionnaire platform to collect subjective scores of a plurality of subjective indicators through the questionnaire platform; collecting electroencephalogram signals according to the effective dominant audio associated with each of the road noise types; performing principal component analysis based on the subjective scores associated with each of the road noise types to determine subjective key principal components corresponding to each of the road noise types; extracting a plurality of spectral features of each of the electroencephalogram signals, performing principal component analysis on the spectral features according to the road noise types to determine objective key principal components corresponding to each of the road noise types; performing comprehensive evaluation using the subjective key principal components, the objective key principal components, the effective dominant audio and the dominant sound pressure levels associated with each of the road noise types to determine an emotional comprehensive index of each of the road noise types. The method comprises the following steps: setting sampling points in one or more functional areas of urban road sound environments to synchronously collect audio samples and time-series sound pressure levels; cutting each of the audio samples at sound event boundaries to determine corresponding sub-audios; classifying each of the sub-audios according to a preset dominant sound source type based on a dominant frequency value to determine a plurality of road dominant sound source types; calculating a dominant frequency confidence interval based on the dominant frequency value of each of the road dominant sound source types to filter out effective dominant frequency audio of each of the road dominant sound source types according to the dominant frequency confidence interval; determining a dominant sound pressure level of each of the effective dominant frequency audios based on the time-series sound pressure levels; The method comprises the following steps: performing average value operation on the dominant frequency values of the effective dominant audio of each of the road noise types and taking a logarithm to determine a corresponding logarithmic dominant frequency value; performing average value operation on the dominant sound pressure levels associated with each of the road noise types to output a corresponding average sound pressure level; performing average value operation on each of the subjective key principal components and each of the objective key principal components to determine a subjective key principal component average value and an objective key principal component average value; performing weighted operation on the subjective key principal component average value, the objective key principal component average value, the logarithmic dominant frequency value and the average sound pressure level to determine an emotional comprehensive index of each of the road noise types.

2. The road sound environment emotional influence evaluation method according to claim 1, characterized by, The method comprises the following steps: performing audio preprocessing on each of the audio samples through Adobe Audition software to output an initialized audio; cutting a plurality of sub-audios according to a Melville spectrogram analysis of each of the initialized audios.

3. The road sound environment emotional influence evaluation method according to claim 1, characterized by, The method comprises the following steps: An interactive front-end page of the questionnaire platform is built based on HTML, CSS and JavaScript; wherein, a page framework is built by HTML, a questionnaire form is designed in combination with subjective indicators, effective dominant audio of each road dominant sound source type is imported based on an Audio component, and an associated audio scene picture is imported based on an image component, a page style is designed by CSS, and page interaction is designed by JavaScript in combination with an Audio component ID; The interactive front-end page and a MySQL database are deployed through a cloud server, a subjective indicator standardization form is established in the MySQL database, Java is used to connect the interactive front-end page and the MySQL database, and the questionnaire platform is determined.

4. The road sound environment emotional influence evaluation method according to claim 1, characterized by, The method further includes: Each of the EEG signals is filtered and artifact removed to output clean EEG signals; According to the power spectrum density of each of the clean EEG signals, the power in different frequency bands is determined as a spectral feature; the frequency bands include Theta waves, Alpha waves and Beta waves.

5. A road sound environment emotion influence evaluation device characterized by comprising: The method further includes: A data sampling module is configured to sample in an urban road sound environment, determine effective dominant audio and a dominant sound pressure level of a plurality of road dominant sound source types, and classify each of the road dominant sound source types into a plurality of road noise types according to a sound source attribute; A subjective evaluation module is configured to embed the effective dominant audio of each of the road dominant sound source types to build a questionnaire platform, and collect subjective scores of a plurality of subjective indicators through the questionnaire platform; An objective evaluation module is configured to trigger the collection of EEG signals according to the associated effective dominant audio of each of the road noise types; A subjective analysis module is configured to perform principal component analysis based on the subjective scores associated with each of the road noise types, respectively, to determine subjective key principal components of each of the road noise types; An objective analysis module is configured to extract a plurality of spectral features of each of the EEG signals, perform principal component analysis on the spectral features according to the road noise types, and determine objective key principal components of each of the road noise types; An emotion evaluation module is configured to perform comprehensive evaluation by using the associated subjective key principal components, objective key principal components, effective dominant audio and dominant sound pressure level, respectively, to determine an emotion comprehensive index of each of the road noise types. The method further includes: In one or more urban road sound environment functional areas, sampling points are set to synchronously collect audio samples and time series sound pressure levels; Each of the audio samples is subjected to sound event boundary cutting to determine corresponding sub-audio; According to the dominant frequency values of each of the sub-audio, the road dominant sound source types are classified according to a preset dominant sound source type, and a plurality of road dominant sound source types are determined; Based on the dominant frequency values of each of the road dominant sound source types, a dominant frequency confidence interval is calculated, and effective dominant frequency audio of each of the road dominant sound source types is filtered according to the dominant frequency confidence interval; Based on each of the time series sound pressure levels, a dominant sound pressure level of the corresponding effective dominant frequency audio is determined. The mood evaluation module is specifically used for: Performing average value operation on the dominant frequency values of the effective dominant audio of each road noise type and taking logarithm to obtain corresponding logarithmic dominant frequency values; Performing average value operation on the associated dominant sound pressure levels according to each road noise type, and outputting corresponding sound pressure level averages; Performing average value operation on each subjective key principal component and each objective key principal component respectively to determine corresponding subjective key principal component averages and objective key principal component averages; Performing weighted operation on the associated subjective key principal component averages, objective key principal component averages, logarithmic dominant frequency values and sound pressure level averages to determine mood comprehensive indexes of each road noise type.

6. A computer device, comprising: The computer program / instructions are executed by the processor to implement the steps of the road sound environment mood influence evaluation method according to any one of claims 1-4.

7. A computer readable storage medium having stored thereon computer programs / instructions, characterized in that, The computer program / instructions are executed by the processor to implement the steps of the road sound environment mood influence evaluation method according to any one of claims 1-4.

8. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instructions are executed by the processor to implement the steps of the road sound environment mood influence evaluation method according to any one of claims 1-4.

Images