阅读说明：本技术 一种基于音频事件检测的高速公路诱导灯系统及方法 (Expressway induction lamp system and method based on audio event detection ) 是由 陈广辉 黄涛 高林 熊斯鹏 吴传洁 黄敏 杨端建 于 2021-05-18 设计创作，主要内容包括：本发明涉及一种基于音频事件检测的高速公路诱导灯系统及方法,所述高速公路诱导灯系统包括安装在道路两侧的诱导灯、边缘计算服务器及控制中心的服务器平台；诱导灯之间、诱导灯系统与边缘计算服务器之间以无线自组网方式通信,用于传输各诱导灯麦克风采集的音频数据；边缘计算服务器与服务器平台之间以移动运营商5G通信方式通信,通过移动运营商的5G通信网络将检测的交通事件和状态上报,并更新音频算法样本库。本发明技术方案除了具有同步闪烁和尾迹跟踪的功能外,在每个诱导灯中内置麦克风,实时采集现场音频；当车辆发生交通事故时产生车辆碰撞声音,通过边缘计算服务器音频AI算法分析,从而检测出交通事故,并将事故位置信息上传至控制中心。(The invention relates to an expressway induction lamp system and a method based on audio event detection, wherein the expressway induction lamp system comprises induction lamps arranged on two sides of a road, an edge calculation server and a server platform of a control center; the induction lamps, the induction lamp system and the edge computing server are communicated in a wireless ad hoc network mode and are used for transmitting audio data collected by each induction lamp microphone; the edge computing server and the server platform communicate in a mobile operator 5G communication mode, the detected traffic events and states are reported through a 5G communication network of the mobile operator, and an audio algorithm sample base is updated. The technical scheme of the invention has the functions of synchronous flicker and trail tracking, and is characterized in that a microphone is arranged in each induction lamp to collect the field audio in real time; when a traffic accident occurs to the vehicle, vehicle collision sound is generated, the vehicle collision sound is analyzed through an audio AI algorithm of the edge computing server, so that the traffic accident is detected, and accident position information is uploaded to the control center.)

1. An induction lamp system for an expressway based on audio event detection comprises induction lamps arranged on two sides of a road, and is characterized by further comprising an edge calculation server and a server platform arranged in a control center; the induction lamps, the induction lamp system and the edge computing server are communicated in a wireless ad hoc network mode and are used for transmitting audio data collected by each induction lamp microphone; the edge computing server communicates with the server platform in a 5G communication mode of a mobile operator, reports the detected traffic events and states through a 5G communication network of the mobile operator, and updates an audio algorithm sample base.

2. The highway induction lamp system based on audio event detection as claimed in claim 1, wherein the induction lamp is distributed on two sides of a road and is powered by solar energy.

3. The highway induction lamp system based on the audio event detection as claimed in claim 2, wherein the induction lamps are dispersedly arranged at the two sides of the highway road at an installation distance of 10-20 m.

4. The highway induction lamp system based on audio event detection as recited in claim 3, wherein a microphone array and an audio acquisition module are arranged in the induction lamp.

5. The highway induction lamp system based on audio event detection as claimed in claim 4, wherein the microphone array collects sound and transmits the sound to the audio collection module, the audio collection module is wirelessly connected with the edge computing server, and the edge computing server is connected with a server platform of the control center through a 5G communication network.

6. The highway induction lamp system based on audio event detection as recited in claim 1, wherein the server platform is provided with a detection result processing module.

7. The highway induction lamp system based on audio event detection as claimed in claim 5, wherein the audio collection module comprises a pre-audio threshold detection circuit, an audio noise suppression circuit, a post-audio threshold detection circuit, an abnormal audio detection initial judgment circuit, an audio data compression circuit, an abnormal audio depth detection circuit and an audio abnormal event reporting circuit.

8. The highway induction lamp system based on audio event detection as claimed in claim 7, wherein a front audio threshold detection circuit of the audio collection module is connected with the microphone array and the audio noise suppression circuit, the rear audio threshold detection circuit, the abnormal audio detection initial judgment circuit, the audio data compression circuit, the abnormal audio depth detection circuit and the audio abnormal event reporting circuit are connected, and the audio abnormal event reporting circuit is connected with the detection result processing module of the server platform through a 5G communication network; the audio acquisition module monitors road surface sound, carries out sound identification to producing vehicle collision sound or sharp-pointed brake sound when taking place the traffic accident, detects whether take place the traffic accident on the road surface to report this incident and the position that takes place initiatively, inform the rescue personnel in time to arrive at the scene and expand rescue work, simultaneously with traffic police, highway information board linkage.

9. A highway audio event detection method using the highway induction lamp system based on audio event detection as claimed in any one of claims 1 to 8, wherein the induction lamp system is composed of induction lamps installed on both sides of a road, an edge computing server, and a server platform deploying a control center, and the highway audio event detection method comprises the following steps:

the induction lamps are powered by solar energy and are arranged on two sides of a road in a distributed manner, each induction lamp is internally provided with an audio acquisition module, a front audio threshold detection circuit of the audio acquisition module is connected with a microphone array and an audio noise suppression circuit, the audio noise suppression circuit, a rear audio threshold detection circuit, an abnormal audio detection initial judgment circuit, an audio data compression circuit, an abnormal audio depth detection circuit and an audio abnormal event reporting circuit are connected, the audio abnormal event reporting circuit is connected with a detection result processing module of a server platform through a 5G communication network to monitor road surface sound, perform sound identification on vehicle collision sound or sharp brake sound generated when a traffic accident occurs, detect whether the traffic accident occurs on the road surface, actively report the event and the position of the event to notify rescue workers to timely rush to the site for rescue work, meanwhile, traffic police, highway information boards and signal systems correspondingly prompt vehicles coming and going.

10. The method for detecting highway audio events according to claim 9, wherein the induction lamps, the induction lamp system and the edge computing server communicate with each other in a wireless ad hoc network manner, and are used for transmitting audio data collected by each induction lamp microphone; the edge computing server communicates with the server platform in a 5G communication mode of a mobile operator, reports the detected traffic events and states through a 5G communication network of the mobile operator, and updates an audio algorithm sample library; the audio event detection detects abnormal sound events such as collision, emergency brake, tire burst and explosion under various road conditions through the sound collection unit arranged on the induction lamps on two sides of the highway by an audio AI algorithm, and comprises the following steps:

a microphone array of an audio acquisition module of the induction lamp acquires road field sounds, energy and frequency information of the sounds are analyzed through a front audio threshold detection module, whether the input sounds are normal sounds or not is judged, and when a detection result exceeds a preset threshold, a subsequent algorithm function is started;

the audio noise suppression circuit is used as a noise reduction module and mainly has the main functions of improving the signal-to-noise ratio of detected sound, eliminating noises which interfere with detection, including wind noise and rainwater noise generated under different weather conditions and road noise generated by running of various vehicles, and improving the detection accuracy;

carrying out primary audio threshold detection on the vehicle running noise exceeding the threshold after noise reduction; the rear audio threshold detection circuit reduces unnecessary calculation, so that the power consumption of the audio acquisition module of the induction lamp is further reduced;

the audio acquisition module of the induction lamp is a micro-control module, an abnormal audio detection initial judgment circuit is arranged, and an algorithm with relatively low calculation complexity is adopted for abnormal sound detection to perform abnormal audio detection initial judgment;

after audio data compression is carried out, sending the audio data to an edge computing server in an induction lamp wireless networking communication mode;

carrying out abnormal audio depth detection on an edge calculation server, deploying one edge calculation server on one road section, calculating and processing audio data collected by all induction lamps in the road section, and carrying out audio depth detection after restoring compressed audio data by adopting a perfect sound detection algorithm by the server;

after the edge computing server detects an abnormal event, reporting the event information to a server platform through a mobile communication network;

and the server platform receives the reported abnormal events, starts an emergency program according to the management strategy, pushes the information to rescue workers and is linked with a traffic police and an expressway information board.

Technical Field

The invention relates to the technical field of highway safety facilities, in particular to a highway induction lamp system based on audio event detection and a highway audio event detection method.

Background

The highway induction lamp system is generally arranged on a highway accident multi-occurrence section, such as a continuous curve, a fog area section and the like, and the induction lamps arranged on guardrails on two sides of a road flicker synchronously, so that the highway induction lamp system has the functions of strengthening the outline and the trend of the road and improving the identification degree of a driver; when the vehicle runs through the induction lamp, the induction lamp at the rear part is turned on to become red, a red trail tracking belt is formed, and the vehicle at the rear part is warned to keep a safe driving distance.

Traffic safety accidents on expressways often occur in rainy and foggy days, and due to visibility, video-based event detection cannot be detected. Because the traffic accident can not be found in time, the rescuers can not arrive at the scene at the first time to carry out the rescue work, and the best rescue opportunity is delayed. However, when a traffic accident occurs, sharp brake sound and vehicle collision sound are usually accompanied, if the sound is identified by the induction lamps arranged on the guardrails on the two sides of the road, the traffic accident is detected, the accident and the occurring position are actively reported, rescue workers are informed to arrive at the scene in time to carry out rescue work, precious time is won for rescue, and the casualty rate of the accident is reduced to the maximum extent.

Disclosure of Invention

The invention aims to provide an expressway induction lamp system based on audio event detection and an expressway audio event detection method. By adopting the detection method, when a vehicle has a traffic accident, vehicle collision sound is generated, the audio AI algorithm of the edge calculation server is used for analyzing, so that the traffic accident is detected, and the accident position information is uploaded to the control center.

In order to achieve the above object, the present invention adopts the following technical solutions.

The invention firstly provides an inducing lamp system for an expressway based on audio event detection, which comprises inducing lamps arranged on two sides of a road, an edge computing server and a server platform arranged in a control center, wherein the inducing lamps are arranged on the two sides of the road; the induction lamps, the induction lamp system and the edge computing server are communicated in a wireless ad hoc network mode and are used for transmitting audio data collected by each induction lamp microphone; the edge computing server communicates with the server platform in a 5G communication mode of a mobile operator, reports the detected traffic events and states through a 5G communication network of the mobile operator, and updates an audio algorithm sample base.

Preferably, the induction lamps are dispersedly installed on two sides of the road and are powered by solar energy.

In any one of the above technical solutions, preferably, the induction lamps are installed at the two sides of the expressway in a dispersed manner at an installation distance of 10-20 m.

In any one of the above technical solutions, preferably, the induction lamp is provided with a microphone array and an audio acquisition module, the microphone array acquires sound and transmits the sound to the audio acquisition module, the audio acquisition module is wirelessly connected to the edge computing server, and the edge computing server is connected to the server platform of the control center through a 5G communication network.

In any of the above technical solutions, preferably, the server platform is provided with a detection result processing module, and the audio acquisition module includes a pre-audio threshold detection circuit, an audio noise suppression circuit, a post-audio threshold detection circuit, an abnormal audio detection initial judgment circuit, an audio data compression circuit, an abnormal audio depth detection circuit, and an audio abnormal event reporting circuit; the system comprises a front audio threshold detection circuit of an audio acquisition module, an audio noise suppression circuit, a rear audio threshold detection circuit, an abnormal audio detection initial judgment circuit, an audio data compression circuit, an abnormal audio depth detection circuit and an audio abnormal event reporting circuit, wherein the front audio threshold detection circuit of the audio acquisition module is connected with a microphone array and the audio noise suppression circuit; the audio acquisition module monitors road surface sound, carries out sound identification to producing vehicle collision sound or sharp-pointed brake sound when taking place the traffic accident, detects whether take place the traffic accident on the road surface to report this incident and the position that takes place initiatively, inform the rescue personnel in time to arrive at the scene and expand rescue work, simultaneously with traffic police, highway information board linkage.

The invention also provides a method for detecting the audio event of the expressway, which adopts the expressway induction lamp system based on the audio event detection as described in any one of the above, wherein the induction lamp system consists of induction lamps arranged at two sides of a road, an edge computing server and a server platform for deploying a control center, and the method for detecting the audio event of the expressway comprises the following steps:

the induction lamps are powered by solar energy and are arranged on two sides of a road in a distributed manner, each induction lamp is internally provided with an audio acquisition module, a front audio threshold detection circuit of the audio acquisition module is connected with a microphone array and an audio noise suppression circuit, the audio noise suppression circuit, a rear audio threshold detection circuit, an abnormal audio detection initial judgment circuit, an audio data compression circuit, an abnormal audio depth detection circuit and an audio abnormal event reporting circuit are connected, the audio abnormal event reporting circuit is connected with a detection result processing module of a server platform through a 5G communication network to monitor road surface sound, perform sound identification on vehicle collision sound or sharp brake sound generated when a traffic accident occurs, detect whether the traffic accident occurs on the road surface, actively report the event and the position of the event to notify rescue workers to timely rush to the site for rescue work, meanwhile, traffic police, highway information boards and signal systems correspondingly prompt vehicles coming and going.

In any of the above technical solutions, preferably, the induction lamps, the induction lamp system and the edge computing server communicate with each other in a wireless ad hoc network manner, and are used for transmitting audio data collected by each induction lamp microphone; the edge computing server communicates with the server platform in a 5G communication mode of a mobile operator, reports the detected traffic events and states through a 5G communication network of the mobile operator, and updates an audio algorithm sample library; the audio event detection detects abnormal sound events such as collision, emergency brake, tire burst and explosion under various road conditions through the sound collection unit arranged on the induction lamps on two sides of the highway by an audio AI algorithm, and comprises the following steps:

a microphone array of an audio acquisition module of the induction lamp acquires road field sounds, energy and frequency information of the sounds are analyzed through a front audio threshold detection module, whether the input sounds are normal sounds or not is judged, and when a detection result exceeds a preset threshold, a subsequent algorithm function is started;

the audio noise suppression circuit is used as a noise reduction module and mainly has the main functions of improving the signal-to-noise ratio of detected sound, eliminating noises which interfere with detection, including wind noise and rainwater noise generated under different weather conditions and road noise generated by running of various vehicles, and improving the detection accuracy;

carrying out primary audio threshold detection on the vehicle running noise exceeding the threshold after noise reduction; the rear audio threshold detection circuit reduces unnecessary calculation, so that the power consumption of the audio acquisition module of the induction lamp is further reduced;

the audio acquisition module of the induction lamp is a micro-control module, an abnormal audio detection initial judgment circuit is arranged, and an algorithm with relatively low calculation complexity is adopted for abnormal sound detection to perform abnormal audio detection initial judgment;

after audio data compression is carried out, sending the audio data to an edge computing server in an induction lamp wireless networking communication mode;

carrying out abnormal audio depth detection on an edge calculation server, deploying one edge calculation server on one road section, calculating and processing audio data collected by all induction lamps in the road section, and carrying out audio depth detection after restoring compressed audio data by adopting a perfect sound detection algorithm by the server;

after the edge computing server detects an abnormal event, reporting the event information to a server platform through a mobile communication network;

and the server platform receives the reported abnormal events, starts an emergency program according to the management strategy, pushes the information to rescue workers and is linked with a traffic police and an expressway information board.

Compared with the prior art, the technical scheme of the invention has the following beneficial effects:

the highway induction lamp system based on the audio event detection is composed of induction lamps arranged on two sides of a road, an edge calculation server and a server platform for deploying a control center, and is simple in structure and easy to implement. The induction lamps, the induction lamp system and the edge computing server are communicated in a wireless ad hoc network mode and are used for transmitting audio data collected by each induction lamp microphone; and the edge computing server and the server platform report the detected traffic events and states through the 5G communication of the mobile operator and update the audio algorithm sample library. The highway induction lamp system based on the audio event detection has the functions of synchronous flicker and trail tracking, and is internally provided with a microphone in each induction lamp to collect field audio in real time. By adopting the highway audio event detection method, when a traffic accident occurs to a vehicle, vehicle collision sound is generated, the vehicle collision sound is analyzed through an audio AI algorithm of the edge computing server, so that the traffic accident is detected, and accident position information is uploaded to a control center.

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, and 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 these drawings without creative efforts.

FIG. 1 is a system diagram of a preferred embodiment of an audio event detection based highway induction light system according to the present invention;

fig. 2 is a schematic workflow diagram of a preferred embodiment of the highway audio event detection method of the highway induction lamp system based on audio event detection according to the invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In order to solve the problem that the highway induction lamp system cannot realize detection based on video event detection under the conditions of poor road conditions and poor visibility in the prior art, the embodiment of the invention provides the highway induction lamp system based on audio event detection and the highway audio event detection method, wherein the induction lamp system has the functions of synchronous flicker and trail tracking, and is internally provided with a microphone in each induction lamp for collecting field audio in real time; by adopting the detection method, when a vehicle has a traffic accident, vehicle collision sound is generated, the audio AI algorithm of the edge calculation server is used for analyzing, so that the traffic accident is detected, and the accident position information is uploaded to the control center.

The following describes the structure and technical features of the highway induction lamp system based on audio event detection and the implementation form of the highway audio event detection method in this embodiment with reference to fig. 1 to 2.

The highway induction lamp system based on audio event detection described in this embodiment includes induction lamps installed on both sides of a road, an edge computing server, and a server platform deployed in a control center, as shown in fig. 1. The induction lamps and the induction lamp system and the edge computing server are communicated in a wireless ad hoc network mode and are used for transmitting audio data collected by each induction lamp microphone. The edge computing server and the server platform communicate in a 5G communication mode of a mobile operator, the detected traffic events and states are reported through a 5G communication network of the mobile operator, and an audio algorithm sample base is updated.

In the highway induction lamp system based on audio event detection described in this embodiment, the induction lamps are installed on both sides of the road in a distributed manner, and the induction lamps are powered by solar energy.

In the expressway induction lamp system based on audio event detection described in this embodiment, the installation intervals of the induction lamps dispersedly installed on both sides of the expressway road are 10-20 meters.

In the highway induction lamp system based on audio event detection described in this embodiment, a microphone array and an audio acquisition module are arranged in the induction lamp.

In the highway induction lamp system based on audio event detection, the microphone array collects sound and transmits the sound to the audio collection module, the audio collection module is wirelessly connected with the edge computing server, and the edge computing server is connected with a server platform of the control center through a 5G communication network.

In the highway induction lamp system based on audio event detection described in this embodiment, the server platform is provided with a detection result processing module.

In the highway induction lamp system based on audio event detection described in this embodiment, the audio acquisition module may include a front audio threshold detection circuit, an audio noise suppression circuit, a rear audio threshold detection circuit, an abnormal audio detection initial judgment circuit, an audio data compression circuit, an abnormal audio depth detection circuit, and an audio abnormal event reporting circuit. A front audio threshold detection circuit of the audio acquisition module is connected with the microphone array and the audio noise suppression circuit, a rear audio threshold detection circuit, an abnormal audio detection initial judgment circuit, an audio data compression circuit, an abnormal audio depth detection circuit and an audio abnormal event reporting circuit are connected, and the audio abnormal event reporting circuit is connected with a detection result processing module of the server platform through a 5G communication network; the audio acquisition module monitors road surface sound, carries out sound identification to producing vehicle collision sound or sharp-pointed brake sound when taking place the traffic accident, detects whether take place the traffic accident on the road surface to report this incident and the position that takes place initiatively, inform the rescue personnel in time to arrive at the scene and expand rescue work, simultaneously with traffic police, highway information board linkage.

The highway audio event detection method adopting the highway induction lamp system based on the audio event detection is implemented in the following form:

the induction lamps are powered by solar energy and are arranged on two sides of a road in a distributed manner, each induction lamp is internally provided with an audio acquisition module, a front audio threshold detection circuit of the audio acquisition module is connected with a microphone array and an audio noise suppression circuit, the audio noise suppression circuit, a rear audio threshold detection circuit, an abnormal audio detection initial judgment circuit, an audio data compression circuit, an abnormal audio depth detection circuit and an audio abnormal event reporting circuit are connected, the audio abnormal event reporting circuit is connected with a detection result processing module of a server platform through a 5G communication network to monitor road surface sound, perform sound identification on vehicle collision sound or sharp brake sound generated when a traffic accident occurs, detect whether the traffic accident occurs on the road surface, actively report the event and the position of the event to notify rescue workers to timely rush to the site for rescue work, meanwhile, traffic police, highway information boards and signal systems correspondingly prompt vehicles coming and going.

The induction lamps, the induction lamp system and the edge computing server are communicated in a wireless ad hoc network mode and are used for transmitting audio data collected by each induction lamp microphone; the edge computing server and the server platform communicate in a 5G communication mode of a mobile operator, the detected traffic events and states are reported through a 5G communication network of the mobile operator, and an audio algorithm sample base is updated. The audio event detection mainly detects abnormal sound events such as collision, emergency braking, tire burst, explosion and the like under various road conditions through audio AI algorithms by sound collection units arranged on induction lamps on two sides of the highway. As shown in fig. 2, in the implementation, the method for detecting an audio event on a highway includes the following steps:

a microphone array of an audio acquisition module of the induction lamp acquires road field sounds; because the induction lamp supplies power for the solar energy and the battery, in order to further reduce the power consumption and the operation amount, an audio threshold detection module is added; the audio threshold detection module is a preposed audio threshold detection module and mainly has the functions of analyzing information such as energy, frequency and the like of sound and judging whether the input sound is normal sound or not; the audio acquisition module of the induction lamp analyzes the energy and frequency information of the sound through the front audio threshold detection module, judges whether the input sound is normal sound, and starts the subsequent algorithm function when the detection result exceeds a preset threshold;

the audio noise suppression circuit is used as a noise reduction module and mainly has the main functions of improving the signal-to-noise ratio of detected sound, eliminating noises which interfere with detection, including wind noise and rainwater noise generated under different weather conditions and road noise generated by running of various vehicles, and improving the detection accuracy;

because the noise of the normal running of some vehicles exceeds the threshold value, the running noise of the vehicles exceeding the threshold value needs to be subjected to audio threshold value detection again after noise reduction, and the post-audio threshold value detection circuit can reduce unnecessary calculation, so that the power consumption of the audio acquisition module of the induction lamp is further reduced;

the abnormal audio detection initial judgment is needed; the audio acquisition module of the induction lamp adopts a micro-control module with low cost, low power consumption and small operand, and the abnormal sound detection adopts an algorithm with relatively low calculation complexity; the detection rate is higher due to the lower initial judgment detection threshold, but the false alarm rate is higher;

after audio data compression is carried out, sending the audio data to an edge computing server in an induction lamp wireless networking communication mode;

carrying out abnormal audio depth detection on an edge calculation server, deploying one edge calculation server on one road section, calculating and processing audio data collected by all induction lamps in the road section, and carrying out audio depth detection after restoring compressed audio data by adopting a perfect sound detection algorithm by the server;

after the edge computing server detects an abnormal event, reporting the event information to a server platform through a mobile communication network;

and the server platform receives the reported abnormal events, starts an emergency program according to the management strategy, pushes the information to rescue workers and is linked with a traffic police and an expressway information board.

In specific implementation, the inducing lamps of the expressway are powered by solar energy and are arranged on two sides of a road in a distributed mode, and the installation distance is generally 10-20 m. This technical scheme adopts to add audio acquisition module in every induction lamp, monitors the road surface sound, produces vehicle collision sound or sharp-pointed brake sound and carries out sound identification when taking place the traffic accident, detects whether take place the traffic accident on the road surface to report this incident and the position of emergence initiatively, inform the first-aid staff in time rush to the scene and expand rescue work, traffic police simultaneously, information board, signal carry out corresponding suggestion. The traditional induction lamp has only a single induction prompt function, is single in function and low in intelligent degree, enables the intelligent degree of the induction lamp system through intelligence of the traditional equipment, and has the characteristics of low cost, convenience in installation and deployment and high recognition rate.

The foregoing is merely illustrative of the preferred embodiments of the present invention and is not to be construed as limiting the scope of the invention; the above description is only for the specific embodiment of the present invention, and is not intended to limit the scope of the present invention; any modification, equivalent replacement, improvement and the like of the technical solution of the present invention by a person of ordinary skill in the art without departing from the design spirit of the present invention shall fall within the protection scope determined by the claims of the present invention.