CN102568153A - Vehicle collision detection device - Google Patents
Vehicle collision detection device Download PDFInfo
- Publication number
- CN102568153A CN102568153A CN2010105952101A CN201010595210A CN102568153A CN 102568153 A CN102568153 A CN 102568153A CN 2010105952101 A CN2010105952101 A CN 2010105952101A CN 201010595210 A CN201010595210 A CN 201010595210A CN 102568153 A CN102568153 A CN 102568153A
- Authority
- CN
- China
- Prior art keywords
- module
- sound
- signal
- dsp module
- dsp
- 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
Images
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention relates to a vehicle collision detection device. Various sound signals are acquired by adopting a sound sensor; after being amplified by an amplification circuit, an electric signal output by the sensor is transmitted into an analog signal input end of a sound acquisition chip; after an analog signal is subjected to A/D (Analog to Digital) conversion by the sound acquisition chip, the converted signal is sent to a DSP (Digital Signal Processor) module to be further processed; and the DSP module processes acquired sound information in real time and judges whether a vehicle collision accident occurs. According to the device, the problem that after the traffic accident occurs, the timely and effective assistance is difficult to implement is solved.
Description
Technical field
The invention belongs to the vehicle security apparatus technical field, more specifically, belong to a kind of collision detecting apparatus for vehicle, belong to automobile-used precision electronic device technical field.
Background technology
The traffic hazard injury becomes a kind of worldwide public hazards that threaten human life's safety day by day.If can in time find the generation and the warning of traffic hazard, can reduce the casualty rate of traffic hazard effectively.
Present stage reduces the traffic hazard injury mainly through using the traffic hazard pick-up unit to realize, mainly be divided into based on magnetic frequently signal, based on spectroscopic signal with based on the vehicle detection apparatus of vision signal.Various pick-up units adopt and comprise that camera, ultrasound wave or microwave wait and detect traffic events; It mainly is the telecommunication flow information of handling macroscopic view; The shortcoming that this indirect detection technique exists that discrimination is not high, time delay is long is difficult to succoured timely and effectively after making traffic hazard take place.
Summary of the invention
For solving the problems of the technologies described above, the invention discloses a kind of collision detecting apparatus for vehicle based on sound detection, can perfectly solve the problems of the technologies described above.
Because traffic hazard can produce very big clash when taking place; And the frequency spectrum of clash and other sound is different; Acoustical signal through gathering and analyze vehicle periphery detects car accident; Can obtain scene of the accident information and warning in real time, therefore good than traffic flow analysis method on instantaneity, and on the recognition success rate of accident, also can improve relatively.Utilize the method for wavelet transformation analysis vehicle sounds to detect car accident, can obtain good recognition effect.Therefore designing a kind of vehicle collision sound detection hardware unit real-time, that accuracy is high has very high practical value.
At first adopt sound transducer to gather various voice signals, the electric signal of sensor output imports the input end of analog signal of sound collection chip into after amplifying circuit amplifies.The sound collection chip is delivered to the DSP module and is done further processing after simulating signal is carried out the A/D conversion.The DSP module is handled the acoustic information that is collected in real time, has judged whether that vehicle collision accident takes place.Memory module links to each other with the DSP module, is used to store the data and the code data of curing that needs are handled, and interim storage space is provided when being the DSP module arithmetic.Alarm module and communication module and external rescue center communicate, in case the DSP module detects the collision happens accident, alarm module just outwards sends warning message.
Vehicle collision warning device of the present invention, volume is little, cost is low.This device uses the mode of dividing frame that acoustical signal is carried out pattern-recognition and calculates, to realize the timely warning of vehicle collision.Experimental result shows that this system reliability is high, time-delay is short, can in time send alerting signal.The application of this system can improve motor vehicles driver and crew's safety coefficient, thereby reduces driver and crew's traffic accident casualty rate, has a good application prospect.
Description of drawings
Fig. 1 is a kind of collision detecting apparatus for vehicle synoptic diagram.Wherein, sound transducer-1; Amplifying circuit-2; Sound collection chip-3; Sound collection module-4; DSP module-5; Report to the police and communication module-6; Memory module-7.
Embodiment
At first adopt sound transducer 1 to gather various voice signals, the electric signal of sensor output imports the input end of analog signal of sound collection chip 3 into after amplifying circuit 2 amplifies.The sound collection chip is delivered to DSP module 5 and is done further processing after simulating signal is carried out the A/D conversion.DSP module 5 is handled the acoustic information that is collected in real time, has judged whether that vehicle collision accident takes place.Memory module 7 links to each other with DSP module 5, is used to store the data and the code data of curing that needs are handled, and interim storage space is provided when being the DSP module arithmetic.Warning and communication module 6 communicate with the external rescue center, in case DSP module 5 detects the collision happens accident, alarm module just outwards sends warning message.
The sound collection module is used the condenser type sound transducer, and SF is 30Hz～18kHz, because the frequency of vehicle noise signal generally can not surpass 10kHz, so this sound transducer can well be realized sampling.Sound transducer is sent into the simulating signal that collects and is sent to the sound collection chip after amplifying circuit amplifies.
The sound collection chip adopts AIC23, and it is a high-performance stereo audio codec chip of TI company, has the 48kHz bandwidth, can satisfy the collection requirement of the voice signal that comprises noise signal.AIC23 carries out stereophony A/D conversion to the simulating signal that collects, and the sampled data of 16bit, 20bit, 24bit and 32bit can be provided under the sampling rate of 8kHz-96kHz.Native system is gathered outside acoustical signal with the 32kHz SF, per second is gathered 32000 voice datas, and the length of image data is made as 16bit, and simulating signal has become 16 digital signal after the A/D conversion like this.Simulating signal is converted into after the digital signal, and AIC23 gives the DSP module with data transmission, supplies the DSP module to do next step processing.
The MODE pin of native system AIC23 is set to O, and control interface is set to the working method of I2C, and what the data transmission interface of AIC23 and DSP module used is the DSP pattern.The DSP module just can be controlled the AIC23 collaborative work like this, and receives the data that AIC23 collects.
The DSP module is the core of total system, accomplish sound signal collection, control, storage, processing and with function such as extraneous communication.That select for use is the dsp chip VC5509 that TI company produces, and it is 16 high fixed DSPs of a cost performance, has a plurality of high performance computations unit, and system clock is 144MHz, and ordering calculation speed is up to 1OOMMACS, and expansion interface on the abundant sheet is provided.
VC5509 has two multichannel buffer serial ports (McBSP), and McBSP has the basic function identical with standard serial interface, and on the basis of standard serial interface, function is expanded.The voice collecting chip AIC23 that native system uses just is connected with DSP through McBSP, and it is as shown in Figure 2 that it connects synoptic diagram.Wherein CLKX is a tranmitting data register, and CLKR is a receive clock, and they all link to each other with the system clock BCLK of AIC23.FSX and FSR realize transmission and the frame synchronization that receives, the LRCIN of corresponding A IC23 and LRCOUT pin.Data are sent pin DX and are linked to each other with DOUT with the DIN of AIC23 respectively with Data Receiving pin DR, accomplish serial data and send and receive operation.
VC5509 also comprises 6 programmable DMA paths, and dma controller can need not the CPU intervention and transmit data on internal storage, external memory storage and the chip between the peripheral hardware, and after operation was accomplished, dma controller can send interrupt request singal to CPU.This system has used a DMA passage 0, reads in data and writes the ad-hoc location of external memory storage from data acquisition module.After data acquisition has been expired, dma controller will produce interruption, and control DSP carries out data processor.The use of DMA has reduced the system break number of times, has obviously improved the travelling speed of system.
Voice data for the sound collection module collects detects software by the operation of DSP module it is analyzed, and judges whether the outside collision has taken place.The DSP module is also drawn an IO mouth and warning and communication module and is communicated, and the signal that warning and communication module receive this pin judges whether to bump to determine whether report to the police.
VC5509 supports unified addressing space, and the total volume of internal memory is 320kB on the sheet, comprises the RAM of 128k * 16bit and the ROM of 32k * 16bit, and can expand to the outer memory headroom of the sheet that is 8M * 16bit to the maximum according to user's needs.What native system was selected for use is the HY57V64 chip, is a SDRAM chip that 4 1M * 16bit logic array is arranged.The acoustical signal data that this chip receives and storage DSP module transmits also can be read corresponding data from the ad-hoc location of this chip when the DSP module needs deal with data.
Alarm module is furnished with GPS and gsm module, obtains position and velocity information and carries out communication with server.The DSP module obtains the speed and the acceleration information of vehicle and adds acoustical signal information to assist calculating from the GPS module of alarm module.Alarm module obtains real-time vehicle collision information from the DSP module, in case detect collision alarm, alarm module is promptly reported to the police to server.
System software is a program that runs on the DSP, each module work of control system, and accomplish algorithm computation.Use the CCS IDE of TI company, programme with C language and assembly language.
This software at first carries out initialization, and the operational factor of VC5509 and AIC23 is configured.During to the phase-locked-loop configuration of VC5509 chip, system clock is set to 144kHz.When McBSP is configured, opens the McBSP0 of VC5509 and start it and carry out input-output operation.Configuration DMA0 passage makes it work in compatibility mode and when interrupting, stops the transmission of data.The mode of operation of configuration AIC23 is the DSP pattern and uses the IIC mode to transmit data.Start AIC23 and acoustical signal is carried out the sampling of 32k sampling rate.
Initialization is carried out sample detecting after finishing, and through sample detecting, satisfies branch frame bar spare in case find the signal that collects, and the acoustical signal length that promptly collects is just carried out automatic sound detection algorithm in the time of enough 1 second.
Automatically sound detection algorithm sense data is also judged, if detected be non-crash events, then continue to carry out sample detecting with etc. pending next second data, at this moment software is being carried out idle loop; When automatic sound detection algorithm is detected is collision accident, just transmits information to communication module, after the GPS module is confirmed speed and positional information, just reports to the police through alarm module.
Because the amplitude versus frequency characte of different acoustic signals is different with phase-frequency characteristic, also there is certain difference in different acoustic signals in the amplitude of each frequency band.Therefore, can utilize the energy changing of each frequency content to realize Target Recognition.
Automatically the sound detection algorithm comprises sound signal collecting and divides frame, feature extraction, characteristic dimensionality reduction, tagsort four parts, and its concrete performing step is following:
(1) collection and branch frame.The signal that collects is divided into a frame by every 2s, the overlapping of 1s is arranged between frame and the frame.Concerning the chip of 32k sampling rate, promptly only 65536 points of fragment of 2s are handled each time, between training stage two fragments, have 1/2 to repeat.Obtain one group of data Datai (1≤i≤65535) like this.
(2) feature extraction.Each frame signal data Datai (1≤i≤65535) is implemented the DWT conversion to obtain frequency domain information, then according to the frequency domain information statistics energy distributions that obtains, with this characteristic as the identification traffic hazard.This algorithm use DB1 small echo to each frame signal, carries out one deck earlier and decomposes, and high frequency coefficient carries out two-layer complete decomposition then, and low frequency coefficient carries out 10 layers of unidirectional decomposition and obtains 18 groups of data.Calculate characteristic component F=[E1; E2 ... E18], the computing formula of En is following:
wherein N is the length of Cn.
(3) characteristic dimensionality reduction.Semaphore after the feature extraction is realized dimensionality reduction.On the basis of the characteristic component F that extracts, this algorithm use detects the traffic hazard clash based on the abnormal point method of determining and calculating of principal component analysis (PCA) (PCA).Obtained after transformation of the original features F
formula
where H is a projection matrix PCA method.
(4) tagsort.Vehicle periphery voice signal sample when collecting normal operation and traffic hazard, and training structural classification device are realized the sound classification in the driving process.Sorter is intended two types of classification results of output: one type is normal operation sound, and another kind of is the collision sound of major traffic accidents.Criterion is:
Wherein
is the characteristic component projection of training sample set.N representes that
allows to deflect away from the maximum number of Ii between the given area; When n is collision during greater than certain threshold value, otherwise then be not collision.
Claims (2)
1. a collision detecting apparatus for vehicle comprises, adopts sound transducer to gather various voice signals; The electric signal of sensor output imports the input end of analog signal of sound collection chip into after amplifying circuit amplifies, after the sound collection chip carries out the A/D conversion with simulating signal; Deliver to the DSP module and do further processing; The DSP module is handled the acoustic information that is collected in real time, has judged whether that vehicle collision accident takes place, and memory module links to each other with the DSP module; Be used to store the data and the code data of curing that needs are handled, and interim storage space is provided when being the DSP module arithmetic.Alarm module and communication module and external rescue center communicate, in case the DSP module detects the collision happens accident, alarm module just outwards sends warning message.
2. the collision detecting apparatus for vehicle described in claim 1 is characterized in that the automatic sound detection algorithm of the acoustic information that the processing in real time of described DSP module is collected comprises sound signal collecting and divides frame, feature extraction, characteristic dimensionality reduction, tagsort four parts.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105952101A CN102568153A (en) | 2010-12-20 | 2010-12-20 | Vehicle collision detection device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010105952101A CN102568153A (en) | 2010-12-20 | 2010-12-20 | Vehicle collision detection device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102568153A true CN102568153A (en) | 2012-07-11 |
Family
ID=46413479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010105952101A Pending CN102568153A (en) | 2010-12-20 | 2010-12-20 | Vehicle collision detection device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102568153A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103745546A (en) * | 2013-12-20 | 2014-04-23 | 青岛海氧益百实业有限公司 | Sound acceleration detection application device used for the fields of communication and security |
| CN103903406A (en) * | 2012-12-30 | 2014-07-02 | 王方淇 | Emergency monitoring and notifying method and system |
| CN104575491A (en) * | 2013-09-30 | 2015-04-29 | 赫拉胡克公司 | Method for recognizing and classifying damage incidents on motor vehicles and device for the same |
| CN104807538A (en) * | 2015-02-12 | 2015-07-29 | 无锡市崇安区科技创业服务中心 | Vehicle collision noise characteristic detection system |
| CN104880245A (en) * | 2015-02-12 | 2015-09-02 | 无锡市崇安区科技创业服务中心 | Vehicle accident positioning and alarming system based on vehicle collision noise characteristics |
| CN105184896A (en) * | 2015-10-08 | 2015-12-23 | 珠海市杰理科技有限公司 | Collision detection device, event data recorder comprising same as well as collision detection processing method |
| CN105486397A (en) * | 2014-10-02 | 2016-04-13 | 赫拉胡克公司 | Sensor device and method for recording at least one contact event on a vehicle |
| CN105632167A (en) * | 2016-03-02 | 2016-06-01 | 苏州华兴源创电子科技有限公司 | Intelligent taxi dispatching system based on mobile terminal |
| CN106056068A (en) * | 2016-05-27 | 2016-10-26 | 大连楼兰科技股份有限公司 | Vehicle low-speed collision signal feature transformation method and system |
| 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 |
| CN108756397A (en) * | 2018-05-04 | 2018-11-06 | 西安工程大学 | A kind of vehicle impact detection system for parking lot |
| WO2019041273A1 (en) * | 2017-08-31 | 2019-03-07 | 深圳市大疆创新科技有限公司 | Impact detection method, impact detection device, and armored vehicle |
| CN110085029A (en) * | 2019-05-31 | 2019-08-02 | 重庆交通大学 | Highway cruising inspection system and method based on rail mounted crusing robot |
| CN111445691A (en) * | 2020-04-03 | 2020-07-24 | 卢思语 | Hidden intersection pedestrian detection, identification and early warning method and system |
| CN112792849A (en) * | 2021-01-06 | 2021-05-14 | 厦门攸信信息技术有限公司 | Collision detection method, robot, mobile terminal and storage medium |
| CN113246894A (en) * | 2020-02-13 | 2021-08-13 | 宁波吉利汽车研究开发有限公司 | Vehicle collision detection method and device |
| CN113246893A (en) * | 2020-02-13 | 2021-08-13 | 宁波吉利汽车研究开发有限公司 | Vehicle collision processing method and system |
| CN113593230A (en) * | 2021-07-27 | 2021-11-02 | 张卫光 | Traffic monitoring system based on wisdom traffic thing networking |
-
2010
- 2010-12-20 CN CN2010105952101A patent/CN102568153A/en active Pending
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103903406A (en) * | 2012-12-30 | 2014-07-02 | 王方淇 | Emergency monitoring and notifying method and system |
| CN104575491B (en) * | 2013-09-30 | 2019-10-18 | 赫拉胡克公司 | The method and device for the method for identification and the damage event on classification vehicle |
| CN104575491A (en) * | 2013-09-30 | 2015-04-29 | 赫拉胡克公司 | Method for recognizing and classifying damage incidents on motor vehicles and device for the same |
| CN103745546A (en) * | 2013-12-20 | 2014-04-23 | 青岛海氧益百实业有限公司 | Sound acceleration detection application device used for the fields of communication and security |
| CN105486397A (en) * | 2014-10-02 | 2016-04-13 | 赫拉胡克公司 | Sensor device and method for recording at least one contact event on a vehicle |
| CN105486397B (en) * | 2014-10-02 | 2020-08-18 | 赫拉胡克公司 | Sensor device and method for detecting at least one contact event at a vehicle |
| CN104807538A (en) * | 2015-02-12 | 2015-07-29 | 无锡市崇安区科技创业服务中心 | Vehicle collision noise characteristic detection system |
| CN104880245A (en) * | 2015-02-12 | 2015-09-02 | 无锡市崇安区科技创业服务中心 | Vehicle accident positioning and alarming system based on vehicle collision noise characteristics |
| CN105184896A (en) * | 2015-10-08 | 2015-12-23 | 珠海市杰理科技有限公司 | Collision detection device, event data recorder comprising same as well as collision detection processing method |
| CN105632167A (en) * | 2016-03-02 | 2016-06-01 | 苏州华兴源创电子科技有限公司 | Intelligent taxi dispatching system based on mobile terminal |
| CN106056068A (en) * | 2016-05-27 | 2016-10-26 | 大连楼兰科技股份有限公司 | Vehicle low-speed collision signal feature transformation method and system |
| WO2019041273A1 (en) * | 2017-08-31 | 2019-03-07 | 深圳市大疆创新科技有限公司 | Impact detection method, impact detection device, and armored vehicle |
| 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 |
| CN108756397A (en) * | 2018-05-04 | 2018-11-06 | 西安工程大学 | A kind of vehicle impact detection system for parking lot |
| CN110085029A (en) * | 2019-05-31 | 2019-08-02 | 重庆交通大学 | Highway cruising inspection system and method based on rail mounted crusing robot |
| CN113246894A (en) * | 2020-02-13 | 2021-08-13 | 宁波吉利汽车研究开发有限公司 | Vehicle collision detection method and device |
| CN113246893A (en) * | 2020-02-13 | 2021-08-13 | 宁波吉利汽车研究开发有限公司 | Vehicle collision processing method and system |
| CN111445691A (en) * | 2020-04-03 | 2020-07-24 | 卢思语 | Hidden intersection pedestrian detection, identification and early warning method and system |
| CN112792849A (en) * | 2021-01-06 | 2021-05-14 | 厦门攸信信息技术有限公司 | Collision detection method, robot, mobile terminal and storage medium |
| CN112792849B (en) * | 2021-01-06 | 2022-07-26 | 厦门攸信信息技术有限公司 | Collision detection method, robot, mobile terminal and storage medium |
| CN113593230A (en) * | 2021-07-27 | 2021-11-02 | 张卫光 | Traffic monitoring system based on wisdom traffic thing networking |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102568153A (en) | Vehicle collision detection device | |
| CN110155064B (en) | Special vehicle driving lane identification and self-vehicle lane change decision-making system and method based on sound signals | |
| CN108898889A (en) | A kind of highway second accident prevention early warning system and control method | |
| CN102945675A (en) | Intelligent sensing network system for detecting outdoor sound of calling for help | |
| WO2005038741A3 (en) | Method and system for collecting traffic data, monitoring traffic, and automated enforcement at a centralized station | |
| CN102103036A (en) | Automobile engine fault diagnosis system based on vibration and audio frequency information | |
| JP2009069885A (en) | State determination device and program | |
| CN103496367A (en) | Method and device for detecting mistaken alarm of lane departure alarming | |
| CN108281002A (en) | The method and system of the retrograde detection of non power driven vehicle based on active RFID | |
| CN102722983A (en) | Audio and video based detection system and method for horn blowing of motor vehicle in violation of regulations | |
| CN111128225A (en) | Vehicle illegal whistle positioning and identifying method | |
| CN104851289B (en) | Traffic comprehensive treatment equipment | |
| CN202362932U (en) | Information monitoring device of traffic accident scene | |
| CN206097466U (en) | Multi -functional intelligent traffic flow monitoring system | |
| CN104880245A (en) | Vehicle accident positioning and alarming system based on vehicle collision noise characteristics | |
| CN107953892A (en) | A kind of vehicle carried driving behavior integration assessment and correction system | |
| CN210258307U (en) | AI intelligent auxiliary driving terminal equipment based on 4G audio and video transmission technology | |
| CN110969850B (en) | Traffic data processing server and method based on secondary processing | |
| CN205609010U (en) | Blind system is led to traffic signal lamp intelligence based on wireless sensor network | |
| CN108269397A (en) | The processing method and mobile unit of a kind of highway traffic data | |
| CN101996492A (en) | Communication method and system for onboard vehicle condition or environmental monitoring information | |
| CN203930967U (en) | A kind of vehicle-mounted polynary running information harvester for the vehicle of insuring | |
| CN217880474U (en) | Perimeter security early warning device of multisource information fusion | |
| CN206179162U (en) | Monitoring of vehicle safe driving and warning platform based on data mining | |
| CN203101409U (en) | DSP (digital signal processor)-based intelligent grain insect detecting device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20120711 |