CN1979575A - Supersonic invasion detection method and detection device - Google Patents
Supersonic invasion detection method and detection device Download PDFInfo
- Publication number
- CN1979575A CN1979575A CN 200510126226 CN200510126226A CN1979575A CN 1979575 A CN1979575 A CN 1979575A CN 200510126226 CN200510126226 CN 200510126226 CN 200510126226 A CN200510126226 A CN 200510126226A CN 1979575 A CN1979575 A CN 1979575A
- Authority
- CN
- China
- Prior art keywords
- signal
- ultrasonic
- echoed signal
- alarm
- invasion
- 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.)
- Granted
Links
Abstract
The invention discloses an ultrasonic invasion detecting method and device. And the method comprises the steps of: (1) receiving echo signal; (2) preprocessing the echo signal; (3) extracting eigenvector from the echo signal; (4) making mode recognition by mode classifier; (5) triggering a warner to give an alarm. And the device comprises: an ultrasonic sounder, an ultrasonic receiver, an A/D converter, a control processor and a warner, where the control processor is composed of pulse generating unit and echo processing unit, the echo processing unit mainly comprises time-domain capturing circuit, frequency-domain filter circuit, characteristic extracting circuit and mode classifier, connected in sequence. And it selects the characteristic in echo signal which is little influenced by external environment, effectively reducing the false alarm rate, and can more flexibly control whether to alarm or not.
Description
Technical field
The present invention relates to a kind of intrusion detection method and sniffer, particularly a kind of supersonic invasion detection method and sniffer based on pattern-recognition.
Background technology
The intrusion detection device is to be used for the moving or the electronics of other action and the device that mechanical part is formed of detecting intrusion person.It comprises initiatively and passive infrared invasion detector, microwave intrusion detector, microwave and the compound intrusion detector of passive infrared, supersonic invasion detector, vibrating intruding detector, sound equipment intrusion detector, magnetic switch intrusion detector, ultrasonic and compound intrusion detector of passive infrared or the like.
Existing intrusion detection device all inevitably can produce false-alarm, in other words conj.or perhaps wrong report.When not having intrusion behavior, the warning that the intrusion detection device sends is called false-alarm.The reason that the intrusion detection device produces false-alarm is a lot, such as element fault or ectocine or the like.The ill effect that false-alarm produced is hardly imaginable, such as because false-alarm has increased many unnecessary troubles the people being fed up with, thereby reduces the confidence level of intrusion detection device greatly.It is contemplated that if at midnight, the managerial personnel in shop and the owner of dwelling house are often waken up with a start owing to false-alarm, they can not be ready to use the high intrusion detection device of this false alarm rate again.The worst consequence is that it unnecessaryly is on the scene police or security personnel at top speed, like this they safety itself and on every side people's safety all can be endangered.Therefore, false-alarm is the deadly defect of existing intrusion detection device.
At present, the microwave Doppler intrusion detection device that extensively exists usually is called as the radar alarm, because it is actually a kind of radar Doppler, use Doppler's principle, radiated electromagnetic wave covers certain investigative range, just reports to the police if detect the object that moves in this scope.Say that technically generally require such intrusion detection device to be made up of one or more sensors and signal processor, sniffer should have the method that can change investigative range.The major defect of microwave intrusion detection device is that installation requirement is higher, if install improperly, microwave signal will penetrate the wall that many windows are housed and cause frequent wrong report.Another shortcoming is that it can send harmful micro-energy, therefore energy must be controlled at harmless level.In addition, microwave intrusion detection device can be subjected to the interference of the used high-energy radar of air traffic and defense sector.
Along with scientific development, the ultrasonic visual field that enters people.Utilize hyperacoustic characteristics and advantage, people have invented the supersonic invasion sniffer.The supersonic invasion sniffer has the following advantages: installs simply, and low energy consumption, low cost, and launch the ultrasound wave that can not listen, and can not alarm the invador, help arresting the invador.
Supersonic Doppler intrusion detection device of the prior art, patent US6 for example, 157,293, patent name is a disclosed technology in " Device for detecting the intrusion of a body in a predetermined space ", its principle and microwave Doppler intrusion detection device are basic identical, and just the signal of emission has become ultrasound wave, utilizes ultrasonic sensor to transmit and receive ultrasonic signal.But there is following shortcoming in this type of ultrasonic intrusion detection device: if it is fine to ventilate in the room, or there is temperature difference in certain position in room, makes air flow bigger, and the ultrasound wave alarm of relative installation is reported by mistake.Because under the bigger situation of air flow, if when transmitting with the wind, it is fast will be more static the time that the ultrasound wave that sends arrives the speed of receiver, and so, the standing wave waveform will be destroyed, thereby trigger alarm.In addition, if the indoor activity that has animals such as dog, cat still can trigger the intrusion detection device, produce higher false alarm rate.
In sum, there is the influence that is subjected to external environment easily in the supersonic invasion sniffer of prior art, produce the deficiency of higher false alarm rate, so people expects a kind of intrusion detection device of low false alarm rate.
Summary of the invention
The objective of the invention is to overcome prior art and have the high deficiency of false alarm rate, thereby propose a kind of accuracy height, intrusion detection method and sniffer that false alarm rate is low based on pattern-recognition.
In order to achieve the above object, the technical scheme taked of the present invention is as follows:
A kind of supersonic invasion detection method comprises the steps:
1) the ultrasound wave acoustical generator is launched ultrasonic pulse signal, and receives echoed signal with ultrasonic receiver;
2) echoed signal is carried out pre-service;
3) echoed signal is extracted proper vector;
4) use pattern classifier to carry out pattern-recognition; If echoed signal belongs to intrusion model, execution in step 5); If echoed signal does not belong to intrusion model, return repeated execution of steps 1);
5) trigger alarm equipment alarm.
In technique scheme, storage assembly intrusion model storehouse in the described pattern classifier of step 4) trains the step of described pattern classifier to comprise: invasion (a) is arranged in enclosure space and not have under two kinds of situations of invasion, gather a plurality of echo signal samples respectively; (b) echo signal samples is carried out pre-service; (c) echo signal samples is extracted proper vector; (d) set minimum accuracy more than 95%,, set up signal intrusion model storehouse by trainable pattern classifier.
In technique scheme, the method for extracting proper vector in the described step (c) comprises the coefficient of choosing after temporal envelope, frequency domain energy or signal pass through conversion (as discrete cosine transform, wavelet transformation etc.).
In technique scheme, further, ultrasound wave acoustical generator described in the step 1) can be the array acoustical generator that piezoelectric ceramics (PZT) transducer or piezoelectric ceramic transducer are formed, or piezoelectric membrane (PVDF) transducer or form the array acoustical generator by piezoelectric thin film transducer, or electrostatic transducer (capacitive transducer) or the array acoustical generator formed by electrostatic transducer, or prevailing ultrasound wave acoustical generator.
In technique scheme, further, receive echoed signal in the step 1) and can use microphone (Microphone) or common ultrasonic sensor, can use a plurality of receptions, also can use single reception.
In technique scheme, further, step 2) described pre-service comprises the time domain intercepting to echoed signal, frequency domain filtering.
In technique scheme, further, step 3) is handled echoed signal, extracts finite dimensional proper vector.
In technique scheme, further, step 4) is carried out pattern-recognition according to the proper vector design category device that extracts in the step 3).Can use neural network classifier (as BP, RBF, SOM network etc.) or other pattern classifiers (as GMM, HMM, SVM etc.).When training classifier, need obtain training sample and test sample book by experiment.The result of identification can simply be divided into two classes: people and unmanned (including the situation of other invaders such as cat, dog) are promptly arranged in the room, also can be divided into three kinds (include the people, thing and unmanned is arranged) and multiple.Obviously, have at sorter under the situation of very high recognition correct rate, system can effectively reduce false alarm rate.
A kind of supersonic invasion sniffer as shown in Figure 1, comprising:
One ultrasound wave acoustical generator 1 is used to launch ultrasonic pulsative signal, and ultrasonic pulsative signal produces echoed signal through 6 emissions of the place ahead barrier;
One ultrasonic receiver 2 is used to receive echoed signal;
One A/D converter 3 is connected with the output terminal of described ultrasonic receiver 2, is used for echoed signal is carried out digital sample;
One processor controls 4 is connected with the input end of described ultrasonic generator 1, the output terminal of described A/D converter 3 and the input end of an alarm 5;
Described alarm 5 is reported to the police after receiving the alerting signal of described processor controls 4.
Further, described processor controls 4 is made up of pulse generation unit and Echo Processing unit; Described pulse generation unit produces the pulse signal that drives described ultrasound wave acoustical generator 1; As shown in Figure 2, described Echo Processing unit is handled echoed signal, comprises that mainly time domain intercepting circuit, frequency domain filtering circuit, characteristic extracting circuit and pattern classifier are linked in sequence.When in described pattern classifier is judged the room invasion being arranged, described Echo Processing unit is exported the alerting signal of a high level and is given described alarm 5, and described alarm 5 is reported to the police; Otherwise, described Echo Processing unit output low level, described alarm is not reported to the police.
Further, described pattern classifier can use neural network classifier (as BP, RBF, SOM network etc.), or other pattern classifiers (as GMM, HMM, SVM etc.).
Further, described ultrasonic receiver 2 can adopt single microphone (Microphone) or single ultrasonic sensor; Also can adopt a plurality of microphones or a plurality of group of ultrasonic sensors to become; If the employing microphone also needs a constant current source to drive this microphone.
Using when of the present invention, ultrasonic generator 1 and receiver 2 can placed a certain angle of enclosure space, so the whole space of covering that investigative range can be as much as possible.After the supersonic invasion sniffer installed, at first to having people and unmanned situation to carry out sample collection in the space, thereby training was used for the sorter of mode identificating circuit.After the sorter training finishes, just can monitor in real time enclosure space.
The present invention considers the acoustic characteristic in room in order to reduce false alarm rate: there is self distinctive sound field in a room, in case there is the invador to enter, the sound field in the room will change, and this change is apparent in view.Therefore utilize the change of this invador of having front and back sound field,, just can provide right judgement by the method for pattern-recognition.If understand, when in the room invador being arranged, can change the generation path of original echo, so the feature of echo will change, and the present invention catches this change exactly, has proposed said method from the angle of echoed signal.
Compared with prior art, the invention has the advantages that:
By the theory of pattern-recognition, to being arranged in the room, the unmanned situation of people distinguishes: choose the very little feature that is affected by the external environment in the echoed signal, effectively reduced false alarm rate.Further, it is that the someone invades on earth that the present invention can also discern in the room, and other animal invasion is still arranged; By the present invention, more deep to signal mode identification, can also discern is owner, or the stranger enters the room.If owner just can not report to the police certainly, otherwise, then report to the police.As seen, the present invention is more flexible to the control of whether reporting to the police.
Description of drawings
Fig. 1 represents the structured flowchart based on the supersonic invasion sniffer of pattern-recognition;
Fig. 2 represents Echo Processing cellular construction synoptic diagram;
Fig. 3 represents the people of a normal room of one embodiment of the invention and the distribution schematic diagram of other invader;
Fig. 4 represents to pass through in one embodiment of the invention the echo signal of intercepting and filtering;
The envelope of the echo signal after the processing when Fig. 5 represents the people is arranged in one embodiment of the invention room with nobody;
Embodiment
Below in conjunction with the drawings and specific embodiments the present invention is described in further detail:
The present invention utilizes the change of invador to the room sound field, and the method by pattern-recognition has effectively reduced false alarm rate.Therefore, need test person and other invader at room diverse location place the change situation to the room sound field, utilize these to record data, trainable pattern classifier, thereby the judged result whether someone invades.
As shown in Figure 1, make a foundation supersonic invasion sniffer based on pattern-recognition of the present invention, comprising: ultrasound wave acoustical generator 1, the emission ultrasonic pulsative signal, ultrasonic signal produces echoed signal through 6 emissions of the place ahead barrier; Ultrasonic receiver 2 receives echoed signal; A/D converter 3 is connected with the output terminal of described ultrasonic receiver 2, and echoed signal is carried out digital sample; Processor controls 4 is connected with the input end of described ultrasonic generator 1, the output terminal of described A/D converter 3 and the input end of an alarm 5; Described alarm 5 is reported to the police after receiving the alerting signal of described processor controls 4.Described processor controls 4 is by pulse generation unit and Echo Processing unit; Described pulse generation unit produces the pulse signal that drives described ultrasound wave acoustical generator 1; Described Echo Processing unit is handled echoed signal, as shown in Figure 2, comprises that mainly time domain intercepting circuit, frequency domain filtering circuit, characteristic extracting circuit and mode identificating circuit are linked in sequence.When the result of described mode identificating circuit judges that man-hour is arranged in the room, described Echo Processing unit is exported the alerting signal of a high level and is given described alarm 5, and described alarm 5 is reported to the police; Otherwise, described Echo Processing unit output low level, described alarm is not reported to the police.Described mode identificating circuit uses BP neural network classifier (hereinafter to be referred as the BP network).
As shown in Figure 3, in closed room (in 4m * 3.5m * 3m), ultrasound wave acoustical generator 1 and receiver 2 place the A place in room, the direction of signal emission is over against the corner on opposite, consider the real-time processing speed to echoed signal, the size in room and the factors such as energy consumption of device, the transmission frequency of ultrasonic transmitter 1 is 40KHz, and length is the sinusoidal signal of 0.5ms, and transmission interval is 10ms.The sample frequency of 3 pairs of echoed signals of A/D converter is 500KHz.
Above each several part circuit, if do not indicate especially, all adopt conventional products well known to those skilled in the art or custom circuit and adopt usual manner to connect.
Owing to need carry out pattern-recognition, according to the supersonic invasion detection method based on pattern-recognition of the present invention, according to following steps:
1) gathers experiment sample, training BP neural network classifier;
Consider objects such as having furniture in the room, in the room, allow people and other invader (choosing an object herein) be in B respectively, C, D, E, five positions of F consider that the size of room and people and object, above-mentioned five positions can cover the situation that people and object in the whole room are in other position basically.The people can be static in the corresponding position or small movements is arranged.The transponder pulse signal, and note echoed signal.100 echoed signals are noted in each position, i.e. 100 samples.The situation that adding had both had no talent in the room when also not having above-mentioned object obtains 11 class samples at last altogether, amounts to 10=1100 sample of (5+5+1) *, respectively each sample is handled, and the process of processing as shown in Figure 2.
At first, as shown in Figure 4, intercept the interior data of preceding 60ms of each echoed signal,, it is given up because the signal after this time is very faint.With the band-pass filter of frequency range at 20～65KHz, filtered echo signal as shown in Figure 4.
Ensuing process is extracted proper vector exactly.Need to prove that the method for selected characteristic vector is a lot, such as choosing temporal envelope, frequency domain energy or signal coefficient after through various conversion (as discrete cosine transform, wavelet transformation etc.) or the like.In this example, because emission is simple signal, echoed signal mainly concentrates on about 40KHz, so do not choose the characteristic quantity relevant with the frequency domain energy, has only chosen the temporal envelope feature; And discover, choose simple temporal envelope feature and can well solve identification problem.The envelope of the echoed signal when Fig. 5 has represented to have in the room people and nobody can significantly be seen the difference of two kinds of envelope signals under the situation.Solid line has represented to have in the room envelope of the echoed signal of man-hour among Fig. 5, and the envelope of the echoed signal when dotted line is represented in the room nobody can see obviously that the dotted line signal Duoed several spikes than solid-line signals.This type of spike is because when not having the invador, the reflection echo of wall produces.This has just illustrated and has extracted envelope signal as the superiority of carrying out the characteristic quantity of pattern-recognition.
The method of extracting proper vector in this example is as follows: at first, calculate the envelope signal of echoed signal, intercept envelope signal with the rectangular window of translation, to the number of signals strong point averaged that at every turn is truncated to and with it as a proper vector.The length (WindowLength) of the length (SignalLength) of the dimension of proper vector (VectorDimension) and signal and rectangular window is relevant like this, promptly
In theory, the big more description to feature of the dimension of proper vector is good more, and the accuracy of identification is high more; But dimension is excessive, also might cause proper vector should not better distinguish in vector space, thereby reduces recognition correct rate; And dimension is big more, and calculated amount is big more, and the training time of BP network is long more.Therefore, the dimension of proper vector is unsuitable excessive, should choose suitable value.This desired value can obtain by testing: as shown in table 1, different dimensions can produce different recognition correct rates.Under the error rate that allows, choose corresponding value.Equal 60 as choosing dimension in second step, the error rate of this moment only has 3.41%.
After proper vector is extracted is exactly BP network.In the every class sample of picked at random 60% formed training sample set, amounts to 660 samples; Remaining 40% as the test sample book collection, amounts to 440 samples.Table 1 has been listed the BP network through the test sample situation after training.
Table 1
| The length of rectangular window | The dimension of proper vector | Recognition correct rate |
| 0.25ms | 240 | 99.31％ |
| 0.5ms | 120 | 97.27％ |
| 1ms | 60 | 96.59％ |
| 2ms | 30 | 88.86％ |
As can be seen from Table 1, along with the reduction of dimension, recognition correct rate is in continuous decline, and this meets the analysis of front fully; On the other hand, the accuracy of identification is also quite high, and first three kind situation has reached more than 96%, and this result has superiority.As previously mentioned, experimental study shows that also can distinguish in the room this moment is someone or the situation that the thing invasion is arranged on earth, and the accuracy of identification has reached more than 94%.
At last, the correct alarm rate that uses the BP network test people train in the room, to walk about.
For better explanation the present invention can use, carried out following test experience under actual conditions.Allow a people in the room, walk about, record 100 groups of samples.Take all factors into consideration the complexity and the accuracy of network training, and the movement velocity of people in the room, the length of choosing rectangular window equals 1ms, and uses the BP network that training finishes in first step experiment under this condition that people and unmanned test are arranged, and the results are shown in table 2.
Table 2
| The length of rectangular window | The dimension of proper vector | Recognition correct rate |
| 1ms | 60 | 95.00％ |
Can see that from the recognition result shown in the table 2 error recognition rate of this invention only has 5% under the actual conditions.In addition, if increase the sample number of BP network, promptly measure the echoed signals at several position place again in the room, the accuracy of the identification of BP network should be able to be higher, thereby obtain lower false alarm rate more.
2) after training BP neural network classifier is finished, carry out actual detection; Ultrasound wave acoustical generator 1 is launched ultrasonic pulse signal, and receives echoed signals with ultrasonic receiver 2;
3) 4 pairs of echoed signals of processor controls are carried out pre-service;
4) 4 pairs of echoed signals of processor controls are extracted proper vector;
5) use pattern classifier to carry out pattern-recognition; If echoed signal belongs to intrusion model, execution in step 5); If echoed signal does not belong to intrusion model, return repeated execution of steps 2);
6) trigger alarm equipment alarm.
After the trigger warning, people can handle accordingly to situation about taking place in the room, handle the work of restarting that later the intrusion detection device returned to form, from above-mentioned steps 2) execution downwards of beginning order.
It should be noted last that above embodiment is only unrestricted in order to technical scheme of the present invention to be described.Although the present invention is had been described in detail with reference to embodiment, those of ordinary skill in the art is to be understood that, technical scheme of the present invention is made amendment or is equal to replacement, do not break away from the spirit and scope of technical solution of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.
Claims (10)
1, a kind of supersonic invasion detection method comprises the steps:
1) the ultrasound wave acoustical generator is launched ultrasonic pulse signal, and receives echoed signal with ultrasonic receiver;
2) echoed signal that step 1) is received is carried out pre-service;
3) pretreated echoed signal is extracted proper vector;
4) use pattern classifier to carry out pattern-recognition; If echoed signal belongs to intrusion model, execution in step 5); If echoed signal does not belong to intrusion model, return repeated execution of steps 1);
5) trigger alarm equipment alarm.
According to the described supersonic invasion detection method of claim 1, it is characterized in that 2, storage assembly intrusion model storehouse in the described pattern classifier of step 4) trains the step of described pattern classifier to comprise:
(a) invasion and not having under two kinds of situations of invasion is arranged in enclosure space, gather a plurality of echo signal samples respectively;
(b) echo signal samples is carried out pre-service;
(c) echo signal samples is extracted proper vector;
(d) set accuracy,, set up signal intrusion model storehouse by trainable pattern classifier.
3, according to claim 1 or 2 described supersonic invasion detection methods, it is characterized in that, the method that described step 3) is extracted proper vector to echoed signal comprises that choosing temporal envelope, frequency domain energy or signal passes through as the coefficient behind discrete cosine transform, the wavelet transformation, extracts finite dimensional proper vector.
4, according to the described supersonic invasion detection method of claim 1, it is characterized in that, ultrasound wave acoustical generator described in the step 1) is meant the array acoustical generator that piezoelectric ceramic transducer or piezoelectric ceramic transducer are formed, or piezoelectric thin film transducer or form the array acoustical generator by piezoelectric thin film transducer, or electrostatic transducer or the array acoustical generator formed by electrostatic transducer, or ultrasound wave acoustical generator.
5, according to the described supersonic invasion detection method of claim 1, it is characterized in that, receive echoed signal in the step 1) and be to use microphone or ultrasonic sensor.
6, according to claim 1 or 2 described supersonic invasion detection methods, it is characterized in that described step 2) in echoed signal carried out pre-service comprise intercepting of the time domain of echoed signal and frequency domain filtering.
7, a kind of supersonic invasion sniffer comprises:
One ultrasound wave acoustical generator (1) is used to launch ultrasonic pulsative signal, and ultrasonic pulsative signal produces echoed signal through barrier (6) emission;
One ultrasonic receiver (2) is used to receive echoed signal;
One A/D converter (3) is connected with the output terminal of described ultrasonic receiver (2), is used for echoed signal is carried out digital sample;
One processor controls (4) is connected with the input end of described ultrasonic generator 1, the output terminal of described A/D converter 3 and the input end of an alarm 5;
Described alarm (5) is reported to the police after receiving the alerting signal of described processor controls (4);
It is characterized in that described processor controls (4) is made up of pulse generation unit and Echo Processing unit; Described pulse generation unit produces the pulse signal that drives described ultrasound wave acoustical generator (1); Described Echo Processing unit is handled echoed signal, comprises that time domain intercepting circuit, frequency domain filtering circuit, characteristic extracting circuit and pattern classifier are linked in sequence, described pattern classifier output signal to described alarm (5).
According to the described supersonic invasion sniffer of claim 7, it is characterized in that 8, described pattern classifier is meant the neural network classifier as BP, RBF, SOM network, or as the pattern classifier of GMM, HMM, SVM.
According to the described supersonic invasion sniffer of claim 7, it is characterized in that 9, described ultrasonic receiver (2) adopts single microphone or a plurality of microphone to form, this intrusion detection device comprises that also a constant current source drives described microphone.
According to claim 7 or 8 described supersonic invasion sniffers, it is characterized in that 10, described ultrasonic receiver (2) adopts single ultrasonic sensor or a plurality of group of ultrasonic sensors to become.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005101262267A CN100440264C (en) | 2005-11-30 | 2005-11-30 | Supersonic invasion detection method and detection device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005101262267A CN100440264C (en) | 2005-11-30 | 2005-11-30 | Supersonic invasion detection method and detection device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1979575A true CN1979575A (en) | 2007-06-13 |
| CN100440264C CN100440264C (en) | 2008-12-03 |
Family
ID=38130725
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005101262267A Expired - Fee Related CN100440264C (en) | 2005-11-30 | 2005-11-30 | Supersonic invasion detection method and detection device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100440264C (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101676963B (en) * | 2008-09-18 | 2011-05-18 | 北京市凯文警视技术研究所 | Invasion body detection alarm equipment and invasion body detection alarm method |
| CN103460263A (en) * | 2011-04-04 | 2013-12-18 | 三菱电机株式会社 | Presence detection system, presence detection method, and program |
| CN103513248A (en) * | 2013-10-23 | 2014-01-15 | 广州市海林电子科技发展有限公司 | Underwater object monitoring device |
| WO2014121627A1 (en) * | 2013-02-05 | 2014-08-14 | 上海联腾通讯科技有限公司 | Signal comparison control unit, anti-theft alarm unit composed thereby, and invisible electronic fence |
| CN104165172A (en) * | 2013-05-16 | 2014-11-26 | 罗伯特·博世有限公司 | Hydraulic device |
| CN105319552A (en) * | 2015-11-26 | 2016-02-10 | 苏州盛世华安智能科技有限公司 | Building anti-climbing monitoring device and building anti-climbing monitoring method |
| US9596549B2 (en) | 2011-01-05 | 2017-03-14 | Koninklijke Philips N.V. | Audio system and method of operation therefor |
| CN108692864A (en) * | 2018-04-17 | 2018-10-23 | 北京汽车集团有限公司 | Parking space closed stratum detection method, system, parking method for early warning and vehicle |
| US10475309B1 (en) | 2018-06-11 | 2019-11-12 | Sray-Tech Image Co., Ltd. | Operation method of smart warning device for security |
| CN110794462A (en) * | 2019-11-06 | 2020-02-14 | 广东博智林机器人有限公司 | Building site safety monitoring system and monitoring method and device thereof |
| CN112215086A (en) * | 2020-09-18 | 2021-01-12 | 轩辕智驾科技（深圳）有限公司 | Method and device for detecting in-vehicle life body signal |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110488278B (en) * | 2019-08-20 | 2021-07-27 | 深圳锐越微技术有限公司 | Doppler radar signal type identification method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2201080C (en) * | 1996-03-27 | 2000-01-25 | Her Majesty The Queen, In Right Of Canada, As Represented By The Ministe R Of Industry | Ultrasonic detection system for safety of vehicle passengers |
| AU751059B2 (en) * | 1998-05-11 | 2002-08-08 | Robert Bosch Gmbh | Ultrasonic monitoring and intruder detection |
| CN2676313Y (en) * | 2003-01-13 | 2005-02-02 | 武汉大学 | Fusion type probe |
| CN2906771Y (en) * | 2005-11-30 | 2007-05-30 | 中国科学院声学研究所 | Ultrasonic wave intrusion detector |
-
2005
- 2005-11-30 CN CNB2005101262267A patent/CN100440264C/en not_active Expired - Fee Related
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101676963B (en) * | 2008-09-18 | 2011-05-18 | 北京市凯文警视技术研究所 | Invasion body detection alarm equipment and invasion body detection alarm method |
| US9596549B2 (en) | 2011-01-05 | 2017-03-14 | Koninklijke Philips N.V. | Audio system and method of operation therefor |
| CN103460263A (en) * | 2011-04-04 | 2013-12-18 | 三菱电机株式会社 | Presence detection system, presence detection method, and program |
| CN103460263B (en) * | 2011-04-04 | 2015-12-09 | 三菱电机株式会社 | In room detection system, at room detection method |
| WO2014121627A1 (en) * | 2013-02-05 | 2014-08-14 | 上海联腾通讯科技有限公司 | Signal comparison control unit, anti-theft alarm unit composed thereby, and invisible electronic fence |
| CN104165172A (en) * | 2013-05-16 | 2014-11-26 | 罗伯特·博世有限公司 | Hydraulic device |
| CN103513248A (en) * | 2013-10-23 | 2014-01-15 | 广州市海林电子科技发展有限公司 | Underwater object monitoring device |
| CN105319552A (en) * | 2015-11-26 | 2016-02-10 | 苏州盛世华安智能科技有限公司 | Building anti-climbing monitoring device and building anti-climbing monitoring method |
| CN108692864A (en) * | 2018-04-17 | 2018-10-23 | 北京汽车集团有限公司 | Parking space closed stratum detection method, system, parking method for early warning and vehicle |
| CN108692864B (en) * | 2018-04-17 | 2020-05-19 | 北京汽车集团有限公司 | Parking space closure degree detection method and system, parking early warning method and vehicle |
| US10475309B1 (en) | 2018-06-11 | 2019-11-12 | Sray-Tech Image Co., Ltd. | Operation method of smart warning device for security |
| TWI680439B (en) * | 2018-06-11 | 2019-12-21 | 視銳光科技股份有限公司 | Operation method of smart warning device for security |
| CN110794462A (en) * | 2019-11-06 | 2020-02-14 | 广东博智林机器人有限公司 | Building site safety monitoring system and monitoring method and device thereof |
| CN110794462B (en) * | 2019-11-06 | 2021-08-03 | 广东博智林机器人有限公司 | Building site safety monitoring system and monitoring method and device thereof |
| CN112215086A (en) * | 2020-09-18 | 2021-01-12 | 轩辕智驾科技（深圳）有限公司 | Method and device for detecting in-vehicle life body signal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100440264C (en) | 2008-12-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100440264C (en) | Supersonic invasion detection method and detection device | |
| CN2906771Y (en) | Ultrasonic wave intrusion detector | |
| CN106054200B (en) | It is a kind of can anti-acoustic jamming ultrasonic distance sensor and its distance measuring method | |
| US5973996A (en) | Ultrasound intrusion detector | |
| Barshan et al. | A bat-like sonar system for obstacle localization | |
| Jung et al. | Echolocation calls in Central American emballonurid bats: signal design and call frequency alternation | |
| US20070052578A1 (en) | Method and system for identifying moving objects using doppler radar | |
| CN103544791A (en) | Underground invasion monitoring system on basis of seismic waves | |
| CN104700543B (en) | Microwave-induced detection method and device used for safety protection | |
| EP3879507A1 (en) | Visual-acoustic monitoring system for event detection, localization and classification | |
| CN103605125A (en) | Method and system for allowing parking sensor to resist interference | |
| EP2464991A1 (en) | A method for human only activity detection based on radar signals | |
| Parker et al. | Indoor target intercept using an acoustic sensor network and dual wavefront path planning | |
| CN112634566B (en) | Intelligent electronic fence construction method and system based on millimeter waves | |
| Beedholm et al. | Automatic gain control in harbor porpoises (Phocoena phocoena)? Central versus peripheral mechanisms | |
| Hammoud et al. | Ultrasense: A self-calibrating ultrasound-based room occupancy sensing system | |
| Barshan | A sonar-based mobile robot for bat-like prey capture | |
| KR101856166B1 (en) | Animal sensing device based on image recognition | |
| CN215123695U (en) | Intelligent driving device | |
| Peremans et al. | Biomimetic platform to study perception in bats | |
| EP3731203A1 (en) | Alarm system | |
| Caicedo et al. | Detection performance analysis of an ultrasonic presence sensor | |
| CN203762132U (en) | Bird repellent device for power transmission line | |
| Li et al. | Ultrasonic intruder detection system for home security | |
| NO341062B1 (en) | Semi-passive transponder |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20081203 Termination date: 20171130 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |