CN112951271A - Tunnel traffic safety warning method and system based on acoustic assistance - Google Patents
Tunnel traffic safety warning method and system based on acoustic assistance Download PDFInfo
- Publication number
- CN112951271A CN112951271A CN202011354836.3A CN202011354836A CN112951271A CN 112951271 A CN112951271 A CN 112951271A CN 202011354836 A CN202011354836 A CN 202011354836A CN 112951271 A CN112951271 A CN 112951271A
- Authority
- CN
- China
- Prior art keywords
- tunnel
- sound
- model
- signal
- traffic safety
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 230000005236 sound signal Effects 0.000 claims abstract description 103
- 238000012549 training Methods 0.000 claims abstract description 42
- 238000013528 artificial neural network Methods 0.000 claims abstract description 9
- 238000006243 chemical reaction Methods 0.000 claims description 15
- 238000001914 filtration Methods 0.000 claims description 14
- 238000012795 verification Methods 0.000 claims description 12
- 230000007246 mechanism Effects 0.000 claims description 11
- 230000003321 amplification Effects 0.000 claims description 10
- 238000003062 neural network model Methods 0.000 claims description 10
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 9
- 230000000694 effects Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 7
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000005728 strengthening Methods 0.000 claims description 6
- 238000013135 deep learning Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 13
- 238000004590 computer program Methods 0.000 description 11
- 238000012544 monitoring process Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 7
- 230000002829 reductive effect Effects 0.000 description 5
- 238000003860 storage Methods 0.000 description 5
- 238000012216 screening Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 206010039203 Road traffic accident Diseases 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000009977 dual effect Effects 0.000 description 2
- 230000002035 prolonged effect Effects 0.000 description 2
- 206010033799 Paralysis Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000002427 irreversible effect Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/091—Traffic information broadcasting
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
- G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Acoustics & Sound (AREA)
- Human Computer Interaction (AREA)
- Evolutionary Computation (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Signal Processing (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a tunnel traffic safety warning method and system based on acoustic assistance, wherein the method comprises the following steps: acquiring sound signals in a tunnel; extracting a characteristic value of a sound signal in the tunnel; inputting the extracted characteristic values into a pre-trained accident type recognition model to obtain accident types; the accident type recognition model is obtained by training a deep neural network based on collected sound signal data in a historical tunnel; and sending out warning information according to the obtained accident type. According to the method, the accident type can be identified by utilizing the pre-trained accident type identification model according to the collected sound signals in the tunnel, so that the condition in the tunnel can be accurately mastered; the accident type can be identified in time and the alarm is given, the attention of the vehicles coming and going is reminded, and the traffic safety is guaranteed.
Description
Technical Field
The invention relates to the technical field of intelligent traffic, in particular to a tunnel traffic safety warning method and system based on acoustic assistance.
Background
The construction of the highway tunnel is closely related to the economic development and the geographic environment of a country or a region. The method has the advantages of shortening mountain-turning and mountain-crossing mileage, saving vehicle transportation time, reducing irreversible damage to natural environment and the like, so the method is widely applied to the geographical environment of mountains in the middle and western parts of China. The space of the road tunnel is nearly closed and the passage is narrow, thereby forming a potential threat to traffic safety to a great extent. According to related data, the traffic accident rate of the road tunnel is far higher than that of the road sections outside the tunnel, and the traffic accident rate in the road tunnel in some areas or provinces reaches more than 3 times of that of the road tunnels outside the tunnel.
The application of the tunnel monitoring system reduces traffic jam and improves the traffic transport network traffic capacity and safety level. However, because the traditional tunnel design concept and the application degree of the modern technology are not complete, the tunnel monitoring system has some defects, such as low intelligent degree, short service life of the monitoring camera in a severe environment, high monitoring cost and the like. With the development of science and technology, the problems of video signal ADIDA conversion damage, multi-equipment backrest transmission loss and the like are solved by means of an intelligent video analysis technology of an intelligent integrated centralized control management platform, and the traditional monitoring is gradually replaced. The following drawbacks still exist: (1) the definition of standard definition video in a dark tunnel environment is poor; (2) the dependence on manual judgment of the accident increases uncertainty; (3) rescue can not be timely obtained after an accident occurs, and the delay of field dredging and cleaning is serious; (4) after an accident occurs, a vehicle behind the accident cannot be informed in time to cause traffic jam and paralysis; (5) accident fire and the 'black hole effect' are easy to cause secondary accidents; (6) the video monitoring maintenance and operation cost is high.
In view of this, there is a need to provide an energy-saving and environment-friendly tunnel monitoring and early warning system capable of making up for the video monitoring defect in tunnel traffic, which is effective, convenient, simple and easy to use.
Chinese utility model patent 201720412366.9 discloses a highway tunnel safety monitoring system, this system sets up a plurality of sound collection module, sound identification module, tunnel portal display screen and information processing and surveillance center in the tunnel. The tunnel portal alarm system has the functions of recognizing sound, collecting sound, alarming on a tunnel portal display screen and transmitting signals to a monitoring center. However, the above patents have the following limitations: (1) 24-hour all-weather monitoring is carried out, the service life of equipment is short, and power resources are wasted; (2) only accidents can be identified, but the type of accident cannot be distinguished.
Disclosure of Invention
The invention provides a tunnel traffic safety warning method and system based on acoustic assistance, and aims to solve the problem that the conventional tunnel safety monitoring system cannot identify the accident type.
In a first aspect, a tunnel traffic safety warning method based on acoustic assistance is provided, which includes:
acquiring sound signals in a tunnel;
extracting a characteristic value of a sound signal in the tunnel;
inputting the extracted characteristic values into a pre-trained accident type recognition model to obtain accident types; the accident type recognition model is obtained by training a deep neural network based on collected sound signal data in a historical tunnel;
and sending out warning information according to the obtained accident type.
Further, the accident type recognition model is obtained by training a deep neural network based on collected historical sound signal data in the tunnel, and comprises the following steps:
acquiring and labeling sound signal data in a historical tunnel;
performing feature extraction on voice signal data in the historical tunnel by using an Auto Model Search mechanism to construct a sample data set, wherein each sample comprises a feature value of a voice signal in the historical tunnel and a corresponding label thereof;
segmenting the sample data set into a training set, a verification set and a test set by utilizing a PaddlePaddle deep learning framework; and carrying out model training on the deep neural network model based on a training set, correcting and strengthening the model by using a verification set, and evaluating the model by using a test set to obtain an accident type identification model.
Further, distributed training is adopted to improve training efficiency in the process of training the deep neural network model based on the training set, and an early termination method (early termination method) mechanism is applied to screen the model in the training process; the overfitting risk can be effectively reduced by applying an earlystopping mechanism, the model screening efficiency is improved, and the training cost is reduced;
the correcting and strengthening of the model by using the verification set comprises the steps of importing sample data of the verification set into a deep neural network model, and verifying the deep neural network model once when a batch (chapter) is trained so as to correct or strengthen model training parameters; the model learning direction can be guided, and the convergence speed is improved;
the evaluation of the model by using the test set comprises the steps of solving the misclassification and constructing a supplementary training set for the misclassified data by using each index of a confusion matrix and F1-score (F1 score, a measure index of a classification problem) by using a model effect evaluation mechanism.
Further, the acquiring the sound signal in the tunnel comprises:
collecting sound signals through sound collecting equipment arranged in the tunnel;
amplifying the sound signal and carrying out filtering processing;
and converting the sound signal after the filtering treatment into a digital quantity signal to obtain a sound signal in the tunnel.
Further, before acquiring the sound signal in the tunnel, the method further includes:
acquiring whether a vehicle enters a tunnel or not, and if so, executing the acquisition of sound signals in the tunnel; otherwise, it is not executed.
In a second aspect, the tunnel traffic safety warning system based on acoustic assistance is provided, and comprises a power supply module, and a sound signal acquisition device, a sound filter, an AD conversion module, a signal identification and classification module and a result output module which are sequentially connected; the sound signal acquisition device, the sound filter, the AD conversion module, the signal identification and classification module and the result output module are all connected with the power supply module;
the sound signal acquisition device is used for acquiring sound signals in the tunnel and transmitting the sound signals to the sound filter;
the sound filter is used for carrying out noise filtration on the received sound signals and transmitting the sound signals to the AD conversion module;
the AD conversion module is used for converting the sound signals after the noise filtration into digital quantity signals to obtain sound signals in the tunnel and transmitting the sound signals to the signal identification and classification module;
the signal identification and classification module is used for receiving sound signals in the tunnel and executing the tunnel traffic safety warning method based on the acoustic assistance to obtain an accident type, and if the accident is confirmed, the accident type is transmitted to the result output module;
and the result output module is used for visually outputting the accident type.
Further, the device also comprises an infrared sensor arranged at the entrance and the exit of the tunnel;
the infrared sensor is connected with the power supply module, and when the infrared sensor detects that a vehicle enters the tunnel, the power supply module supplies power to the warning system; and when the infrared sensor detects that no vehicle exists in the tunnel, the power supply module cuts off power supply.
Furthermore, the sound signal acquisition device comprises a microphone pickup and a signal amplification chip connected with the microphone pickup; the microphone pickup is used for collecting sound signals in the tunnel and transmitting the sound signals to the signal amplification chip; the signal amplification chip is used for amplifying the sound signal of the microphone pickup.
Furthermore, the sound filter is a band-pass filter, the pass band is 300 Hz-3.4 kHz, and the sound filter is used for removing clutter signals.
Furthermore, the result output module comprises a sound device arranged in the tunnel, a display screen arranged at the entrance and the exit of the tunnel and an intelligent alarm device;
the sound equipment is used for broadcasting accident information;
the display screen is used for displaying warning information;
the intelligent alarm device is used for automatically alarming when an accident occurs and transmitting accident information to related departments.
In a third aspect, a computer-readable storage medium is provided, which stores a computer program that, when loaded by a processor, performs the tunnel traffic safety warning method based on acoustic assistance as described above.
Advantageous effects
The invention provides a tunnel traffic safety warning method and system based on acoustic assistance, which have the following advantages:
1. the invention can identify the accident type according to the collected sound signals in the tunnel, thereby being convenient for accurately mastering the condition in the tunnel;
2. the invention can identify the accident type in time and warn to remind the coming and going vehicles to pay attention so as to ensure the traffic safety;
3. whether a vehicle enters the tunnel or not is identified in real time, the vehicle is started or not, and the vehicle is not closed, so that resources are saved;
4. the recognition accuracy can be improved by filtering the noise by filtering the sound signal.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flowchart of a tunnel traffic safety warning method based on acoustic assistance according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a tunnel traffic safety warning system based on acoustic assistance according to an embodiment of the present invention;
FIG. 3 is a diagram of a portion of an acquisition amplifying circuit in the sound signal acquisition device in the embodiment provided in FIG. 2;
FIG. 4 is a circuit diagram of an acoustic filter in the embodiment provided in FIG. 2;
fig. 5 is a flowchart of a tunnel traffic safety warning system based on acoustic assistance according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the examples given herein without any inventive step, are within the scope of the present invention.
As shown in fig. 1, in an aspect of the embodiment of the present invention, there is provided a tunnel traffic safety warning method based on acoustic assistance, including:
s01: acquiring sound signals in a tunnel;
s02: extracting a characteristic value of a sound signal in the tunnel;
s03: inputting the extracted characteristic values into a pre-trained accident type recognition model to obtain accident types; the accident type recognition model is obtained by training a deep neural network based on collected sound signal data in a historical tunnel;
s04: and sending out warning information according to the obtained accident type.
In this embodiment, the training of the deep neural network by the accident type recognition model based on the collected sound signal data in the historical tunnel includes:
acquiring and labeling sound signal data in a historical tunnel; the acquiring of the sound signal data in the historical tunnel specifically comprises: through sound signal data in a history tunnel of the recording equipment, the interference of a system is reduced to the maximum extent, and the validity of the data is ensured; then screening the data, and removing the part with larger noise;
performing feature extraction on voice signal data in the historical tunnel by using an Auto Model Search mechanism to construct a sample data set, wherein each sample comprises a feature value of a voice signal in the historical tunnel and a corresponding label thereof;
segmenting the sample data set into a training set, a verification set and a test set by utilizing a PaddlePaddle deep learning framework; and carrying out model training on the deep neural network model based on a training set, correcting and strengthening the model by using a verification set, and evaluating the model by using a test set to obtain an accident type identification model.
The method comprises the following steps that distributed training is adopted to improve training efficiency in the process of performing model training on the deep neural network based on a training set, and an early stopping method (early stopping method) mechanism is applied to screening the model in the training process; the overfitting risk can be effectively reduced by applying an earlystopping mechanism, the model screening efficiency is improved, and the training cost is reduced;
the correcting and strengthening of the model by using the verification set comprises the steps of importing sample data of the verification set into a deep neural network model, and verifying the deep neural network model once when a batch (chapter) is trained so as to correct or strengthen model training parameters; the model learning direction can be guided, and the convergence speed is improved;
the evaluation of the model by using the test set comprises the steps of solving the misclassification and constructing a supplementary training set for the misclassified data by using each index of a confusion matrix and F1-score (F1 score, a measure index of a classification problem) by using a model effect evaluation mechanism.
In this embodiment, before acquiring the sound signal in the tunnel, the method includes:
collecting sound signals through sound collecting equipment arranged in the tunnel;
amplifying the sound signal and carrying out filtering processing;
and converting the sound signal after the filtering treatment into a digital quantity signal to obtain a sound signal in the tunnel.
Preferably, before acquiring the sound signal in the tunnel, the method further includes:
acquiring whether a vehicle enters a tunnel or not, and if so, executing the acquisition of sound signals in the tunnel; otherwise, it is not executed.
The method for extracting the feature value of the sound signal in the tunnel in step S02 is the same as the feature extraction method in the accident type recognition model training process, and is not described herein again.
In addition, the training of the accident type recognition model is realized based on a PaddlePaddle deep learning framework, namely, the accident type recognition model is obtained by adopting artificial intelligence distributed model training. Because the artificial intelligence training accident type recognition model is adopted, the acquired data and the prediction result data can be acquired in real time subsequently to continuously and autonomously learn and perfect the model, and the recognition precision is improved in real time.
As shown in fig. 2, in one aspect of the embodiment of the present invention, an acoustic-assistance-based tunnel traffic safety warning system is provided, which includes a power module 6, and a sound signal collecting device 1, a sound filter 2, an AD conversion module 3, a signal identifying and classifying module 4, and a result output module 5, which are connected in sequence; the sound signal acquisition device 1, the sound filter 2, the AD conversion module 3, the signal identification and classification module 4 and the result output module 5 are all connected with the power supply module 6;
the sound signal collecting device 1 is used for collecting sound signals in a tunnel and transmitting the sound signals to the sound filter 2;
the sound filter 2 is used for performing noise filtering on the received sound signal and transmitting the sound signal to the AD conversion module 3;
the AD conversion module 3 is used for converting the sound signals after noise filtration into digital quantity signals to obtain sound signals in the tunnel and transmitting the sound signals to the signal identification and classification module 4;
the signal identification and classification module 4 is used for receiving sound signals in the tunnel and executing the tunnel traffic safety warning method based on the acoustic assistance to obtain an accident type, and if the accident is confirmed, the accident type is transmitted to the result output module 5;
and the result output module 5 is used for visually outputting the accident type.
Preferably, the system further comprises an infrared sensor 7 arranged at the entrance and the exit of the tunnel;
the infrared sensor 7 is connected with the power supply module 6, and when the infrared sensor 7 detects that a vehicle enters a tunnel, the power supply module 6 supplies power to the warning system; when the infrared sensor 7 detects that no vehicle exists in the tunnel, the power supply module 6 cuts off power supply. The overall service life of the system can be prolonged, and the service life can be prolonged; the waste of blank period is avoided, and the power resource is saved.
In this embodiment, the sound signal collecting device 1 includes a microphone pickup and a signal amplifying chip connected to the microphone pickup; the microphone pickup is used for collecting sound signals in the tunnel and transmitting the sound signals to the signal amplification chip; the signal amplification chip is used for amplifying the sound signal of the microphone pickup. In specific implementation, the NE5532 signal amplification chip, which is a high-performance low-noise dual operational amplifier (dual operational amplifier) integrated circuit, may be used as the signal amplification chip, and a partial diagram of the acquisition and amplification circuit of the sound signal acquisition device is shown in fig. 3. The high-voltage power supply has the characteristics of better noise resistance, excellent output driving capability, quite high small signal bandwidth, large power supply voltage range and the like. In order to ensure that the sound signal collecting device can collect all sound signals in the tunnel, one sound signal collecting device is arranged at intervals of 20-80 meters, and preferably 50 meters.
In this embodiment, the sound filter 2 is a band pass filter, the pass band is 300Hz to 3.4kHz, and is used for removing clutter signals, and a circuit diagram thereof is shown in fig. 4. The AD conversion module 3 adopts a high-precision ADS1256 chip, and has the characteristics of 8-channel input (8-channel signal input is acquired simultaneously), wide measurement range (the basic range is 0-5V input voltage, a divider resistor can be welded at the input end to adjust the voltage to be within 0-5V), high acquisition frequency and precision (the acquisition rate of an acquisition card is 30K/s, and the precision can reach 0.00001), strong anti-interference capability, small volume, convenience in installation and application, jump cap design and convenience in multi-state combination.
In this embodiment, the result output module 5 includes a sound device disposed in the tunnel, a display screen disposed at the entrance and exit of the tunnel, and an intelligent alarm device;
the sound equipment is used for broadcasting accident information; in order to ensure the broadcasting effect in the tunnel, a sound box is arranged at intervals of 20-80 meters, preferably 50 meters;
the display screen is used for displaying warning information; in order to ensure the early warning effect, the display screen is arranged 50-200 meters in front of the tunnel entrance;
the intelligent alarm device is used for automatically alarming when an accident happens and transmitting accident information to relevant departments, wherein the accident information comprises accident types, occurrence time and places. Optionally, the accident information further includes the location of the accident occurrence location within the tunnel, and each sound signal collecting device has a unique tag, and since the location of each sound signal collecting device is known, the location of the accident occurrence location within the tunnel can be determined by the sound signal collecting device.
The working principle of the tunnel traffic safety warning system is as follows: an accident type recognition model trained in advance is stored in a signal recognition and classification module 4 (a single chip microcomputer can be selected); the microphone pickup collects sound signals, firstly amplifies the sound signals, then filters waves to remove noise, reserves useful audio signals, and finally converts the audio signals from analog quantity to digital quantity through the AD conversion module 3 and transmits the digital quantity to the signal identification and classification module; the signal identification and classification module 4 firstly extracts the characteristic value of the sound signal, and then inputs the characteristic value into a pre-trained accident type identification model to obtain an accident type; the accident types comprise no accident, collision among vehicles, scratch among vehicles, collision between vehicles and the like; and when the accident is identified and confirmed, the accident type is transmitted to the result output module so as to carry out early warning. The specific workflow can be seen in fig. 5.
In another aspect of the embodiments of the present invention, a computer-readable storage medium is provided, which stores a computer program, and when the computer program is loaded by a processor, the computer program executes the tunnel traffic safety warning method based on acoustic assistance.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It is understood that the same or similar parts in the above embodiments may be mutually referred to, and the same or similar parts in other embodiments may be referred to for the content which is not described in detail in some embodiments.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (10)
1. A tunnel traffic safety warning method based on acoustic assistance is characterized by comprising the following steps:
acquiring sound signals in a tunnel;
extracting a characteristic value of a sound signal in the tunnel;
inputting the extracted characteristic values into a pre-trained accident type recognition model to obtain accident types; the accident type recognition model is obtained by training a deep neural network based on collected sound signal data in a historical tunnel;
and sending out warning information according to the obtained accident type.
2. The tunnel traffic safety warning method based on acoustic assistance according to claim 1, wherein the accident type recognition model is obtained by training a deep neural network based on collected historical sound signal data in a tunnel, and comprises the following steps:
acquiring and labeling sound signal data in a historical tunnel;
performing feature extraction on the voice signal data in the historical tunnel by using an Auto Model Search mechanism to construct a sample data set, wherein each sample comprises a feature value of the voice signal in the historical tunnel and a corresponding label thereof;
dividing the sample data set into a training set, a verification set and a test set by using a PaddlePaddle deep learning framework; and carrying out model training on the deep neural network model based on a training set, correcting and strengthening the model by using a verification set, and evaluating the model by using a test set to obtain an accident type identification model.
3. The tunnel traffic safety warning method based on acoustic assistance as claimed in claim 2, wherein distributed training is adopted in the process of model training of the deep neural network based on the training set, and the model is screened by using an early slope mechanism in the training process;
the method for correcting and strengthening the model by using the verification set comprises the steps of importing sample data of the verification set into a deep neural network model, and verifying the deep neural network model once when one batch is trained so as to correct or strengthen model training parameters;
the evaluation of the model by using the test set comprises the steps of solving error classification and constructing a supplementary training set for data of the error classification by using a model effect evaluation mechanism and using each index of a confusion matrix and F1-score.
4. The method for tunnel traffic safety warning based on acoustic assistance according to claim 1, wherein the acquiring of the sound signal in the tunnel comprises:
collecting sound signals through sound collecting equipment arranged in the tunnel;
amplifying the sound signal and carrying out filtering processing;
and converting the sound signal after the filtering treatment into a digital quantity signal to obtain a sound signal in the tunnel.
5. The method for warning of tunnel traffic safety based on acoustic assistance according to claim 1, wherein the step of obtaining the sound signal in the tunnel further comprises:
acquiring whether a vehicle enters a tunnel or not, and if so, executing the acquisition of sound signals in the tunnel; otherwise, it is not executed.
6. A tunnel traffic safety warning system based on acoustic assistance is characterized by comprising a power supply module, and a sound signal acquisition device, a sound filter, an AD conversion module, a signal identification and classification module and a result output module which are sequentially connected; the sound signal acquisition device, the sound filter, the AD conversion module, the signal identification and classification module and the result output module are all connected with the power supply module;
the sound signal acquisition device is used for acquiring sound signals in the tunnel and transmitting the sound signals to the sound filter;
the sound filter is used for carrying out noise filtration on the received sound signals and transmitting the sound signals to the AD conversion module;
the AD conversion module is used for converting the sound signals after the noise filtration into digital quantity signals to obtain sound signals in the tunnel and transmitting the sound signals to the signal identification and classification module;
the signal identification and classification module is used for receiving sound signals in a tunnel and executing the tunnel traffic safety warning method based on the acoustic assistance according to any one of claims 1 to 3 to obtain an accident type, and if the accident is confirmed, the accident type is transmitted to the result output module;
and the result output module is used for visually outputting the accident type.
7. The tunnel traffic safety warning system based on acoustic assistance of claim 6, further comprising an infrared sensor disposed at a tunnel entrance;
the infrared sensor is connected with the power supply module, and when the infrared sensor detects that a vehicle enters the tunnel, the power supply module supplies power to the warning system; and when the infrared sensor detects that no vehicle exists in the tunnel, the power supply module cuts off power supply.
8. The tunnel traffic safety warning system based on acoustic assistance as claimed in claim 6, wherein the sound signal collecting device comprises a microphone pickup and a signal amplifying chip connected with the microphone pickup; the microphone pickup is used for collecting sound signals in the tunnel and transmitting the sound signals to the signal amplification chip; the signal amplification chip is used for amplifying the sound signal of the microphone pickup.
9. The acoustic-assistance-based tunnel traffic safety warning system according to claim 6, wherein the sound filter is a band-pass filter, and the pass band is 300 Hz-3.4 kHz.
10. The tunnel traffic safety warning system based on acoustic assistance as claimed in claim 6, wherein the result output module comprises a sound device arranged in the tunnel, a display screen arranged at the entrance and exit of the tunnel, and an intelligent alarm device;
the sound equipment is used for broadcasting accident information;
the display screen is used for displaying warning information;
the intelligent alarm device is used for automatically alarming when an accident occurs and transmitting accident information to related departments.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011354836.3A CN112951271A (en) | 2020-11-27 | 2020-11-27 | Tunnel traffic safety warning method and system based on acoustic assistance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011354836.3A CN112951271A (en) | 2020-11-27 | 2020-11-27 | Tunnel traffic safety warning method and system based on acoustic assistance |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112951271A true CN112951271A (en) | 2021-06-11 |
Family
ID=76234702
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011354836.3A Pending CN112951271A (en) | 2020-11-27 | 2020-11-27 | Tunnel traffic safety warning method and system based on acoustic assistance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112951271A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113596368A (en) * | 2021-07-23 | 2021-11-02 | 深圳市警威警用装备有限公司 | Information acquisition method based on law enforcement recorder and law enforcement recorder |
| CN114779795A (en) * | 2022-06-21 | 2022-07-22 | 山东金宇信息科技集团有限公司 | Accident dredging method, equipment and medium based on rail robot |
| CN116161051A (en) * | 2022-12-27 | 2023-05-26 | 小米汽车科技有限公司 | Warning method, device, equipment, medium and vehicle for vehicle driver |
| CN118629217A (en) * | 2024-08-14 | 2024-09-10 | 四川智慧高速科技有限公司 | Highway voice recognition accident detection system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104869701A (en) * | 2015-04-27 | 2015-08-26 | 南通古桥新能源科技有限公司 | Street lamp illumination control system |
| CN107086036A (en) * | 2017-04-19 | 2017-08-22 | 杭州派尼澳电子科技有限公司 | A kind of freeway tunnel method for safety monitoring |
| CN108519149A (en) * | 2018-03-28 | 2018-09-11 | 长安大学 | A kind of tunnel accident monitor and alarm system and method based on sound Time-Frequency Analysis |
| CN109346103A (en) * | 2018-10-30 | 2019-02-15 | 交通运输部公路科学研究所 | A kind of audio-frequency detection for highway tunnel traffic event |
| CN109361617A (en) * | 2018-09-26 | 2019-02-19 | 中国科学院计算机网络信息中心 | A kind of convolutional neural networks traffic classification method and system based on network payload package |
| CN209445198U (en) * | 2019-03-21 | 2019-09-27 | 苏州源泽光电科技集团有限公司 | A kind of hanging-ring type intelligent radio remote control switch light modulation bulkhead lamp |
| CN110718234A (en) * | 2019-09-02 | 2020-01-21 | 江苏师范大学 | Acoustic scene classification method based on semantic segmentation coding and decoding network |
-
2020
- 2020-11-27 CN CN202011354836.3A patent/CN112951271A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104869701A (en) * | 2015-04-27 | 2015-08-26 | 南通古桥新能源科技有限公司 | Street lamp illumination control system |
| CN107086036A (en) * | 2017-04-19 | 2017-08-22 | 杭州派尼澳电子科技有限公司 | A kind of freeway tunnel method for safety monitoring |
| CN108519149A (en) * | 2018-03-28 | 2018-09-11 | 长安大学 | A kind of tunnel accident monitor and alarm system and method based on sound Time-Frequency Analysis |
| CN109361617A (en) * | 2018-09-26 | 2019-02-19 | 中国科学院计算机网络信息中心 | A kind of convolutional neural networks traffic classification method and system based on network payload package |
| CN109346103A (en) * | 2018-10-30 | 2019-02-15 | 交通运输部公路科学研究所 | A kind of audio-frequency detection for highway tunnel traffic event |
| CN209445198U (en) * | 2019-03-21 | 2019-09-27 | 苏州源泽光电科技集团有限公司 | A kind of hanging-ring type intelligent radio remote control switch light modulation bulkhead lamp |
| CN110718234A (en) * | 2019-09-02 | 2020-01-21 | 江苏师范大学 | Acoustic scene classification method based on semantic segmentation coding and decoding network |
Non-Patent Citations (1)
| Title |
|---|
| 贺欢欢 等: "基于EasyDL平台视频事件监测系统在铁路运维中的应用", 无线互联科技, vol. 17, no. 08, pages 154 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113596368A (en) * | 2021-07-23 | 2021-11-02 | 深圳市警威警用装备有限公司 | Information acquisition method based on law enforcement recorder and law enforcement recorder |
| CN114779795A (en) * | 2022-06-21 | 2022-07-22 | 山东金宇信息科技集团有限公司 | Accident dredging method, equipment and medium based on rail robot |
| CN114779795B (en) * | 2022-06-21 | 2022-09-20 | 山东金宇信息科技集团有限公司 | Accident dredging method, equipment and medium based on rail robot |
| CN116161051A (en) * | 2022-12-27 | 2023-05-26 | 小米汽车科技有限公司 | Warning method, device, equipment, medium and vehicle for vehicle driver |
| CN116161051B (en) * | 2022-12-27 | 2024-10-18 | 小米汽车科技有限公司 | Warning method, device, equipment, medium and vehicle for vehicle driver |
| CN118629217A (en) * | 2024-08-14 | 2024-09-10 | 四川智慧高速科技有限公司 | Highway voice recognition accident detection system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112951271A (en) | Tunnel traffic safety warning method and system based on acoustic assistance | |
| CN103258430B (en) | Road traveling time calculating and traffic road condition judging method and road traveling time calculating and traffic road condition judging device | |
| WO2018058958A1 (en) | Road vehicle traffic alarm system and method therefor | |
| CN103985182B (en) | A kind of bus passenger flow automatic counting method and automatic counter system | |
| CN102881162A (en) | Data processing and fusion method for large-scale traffic information | |
| CN103150900A (en) | Traffic jam event automatic detecting method based on videos | |
| CN106846803A (en) | Traffic incidents detection device and method based on audio | |
| CN107103775A (en) | A kind of road quality detection method calculated based on gunz | |
| CN101783074A (en) | Method and system for real-time distinguishing traffic flow state of urban road | |
| CN103345842A (en) | Road vehicle classification system and method | |
| CN105489019A (en) | Model-based traffic flow monitoring system based on dual audio signal acquisition | |
| CN109325409B (en) | Passing vehicle fake plate inspection platform | |
| CN104794906A (en) | Vehicle management platform of outdoor parking lot exit | |
| CN111754786A (en) | System for identifying traffic vehicle passing events on highway | |
| CN115797403A (en) | Traffic accident prediction method and device, storage medium and electronic device | |
| CN202422420U (en) | Illegal parking detection system based on video monitoring | |
| CN111274886A (en) | Deep learning-based pedestrian red light violation analysis method and system | |
| CN114973659A (en) | Method, device and system for detecting indirect event of expressway | |
| CN116168356A (en) | Vehicle damage judging method based on computer vision | |
| CN116631186A (en) | Expressway traffic accident risk assessment method and system based on dangerous driving event data | |
| CN111986483A (en) | Method and device for studying and judging road congestion state based on electric alarm data collision and storage medium | |
| CN102622883A (en) | Method and device for judging release of traffic incident | |
| CN103198707A (en) | Vehicle regulation and control method based on dangerous traffic flow characteristics under sunny condition | |
| CN107730882B (en) | Road congestion prediction system and method based on artificial intelligence | |
| CN113156913A (en) | ABS fault diagnosis system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210611 |