CN202404694U - Adaptive disturbance signal identification module of distributing type optical fiber sensing application system - Google Patents
Adaptive disturbance signal identification module of distributing type optical fiber sensing application system Download PDFInfo
- Publication number
- CN202404694U CN202404694U CN2011203607287U CN201120360728U CN202404694U CN 202404694 U CN202404694 U CN 202404694U CN 2011203607287 U CN2011203607287 U CN 2011203607287U CN 201120360728 U CN201120360728 U CN 201120360728U CN 202404694 U CN202404694 U CN 202404694U
- Authority
- CN
- China
- Prior art keywords
- module
- signal
- denoising
- discrimination
- characteristic
- 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.)
- Withdrawn - After Issue
Links
Images
Landscapes
- Alarm Systems (AREA)
Abstract
The utility model relates to an adaptive disturbance signal identification apparatus of a wide-range optical fiber security system. Based on a distributing type all-fiber reflected interference sensing system, according to the generation mechanism and characteristics of various disturbance signals in an environment, the integrated signal processing methods are applied such as a spectral subtraction denoising or wavelet domain denoising, wavelet transformation energy threshold feature extraction and artificial intelligence SVM, the signal classes are identified, and a dangerous event can be accurately identified in time. The realization process comprises content in two aspects as following: first, signal denoising is carried out through the signal denoising module, terminal detection processing is carried out by a terminal detection module, then feature extraction is carried out by a feature extraction module, finally, the input signal class and property are identified by the SVM identification module; and second, whether the identified class of the input signal is displayed is determined by a main controller. According to the utility model, accurate signal identification is provided for building peripheral safety monitor of an important enforcement and an area, and safety guard is provided.
Description
Technical field
The utility model relates to a kind of signal identification module, and especially a kind of distributing optical fiber sensing application system self-adaptation type disturbing signal identification module belongs to the technical field that signal is discerned.
Background technology
Along with science and technology and social development, the more and more fierce that becomes of the competition between each countries and regions, meanwhile, various dangerous terrorist activities are also more and more fiery.Therefore, for some critical facilitys or zone, like military restricted zone, high-risk forbidden zone and airport, nuclear power station or the like needs highly sensitive, intelligentized safety monitoring system to guarantee the safety in these facilities or zone.At present; The technology that is used for the security protection of buildings circumference has infrared, microwave and traditional technology such as iron wire electrical network, but they all have fatal separately shortcoming; Be difficult to signal is carried out accurate complex processing, therefore to the circumference safeguard protection, forbid that the efficient of illegal invasion is lower.Investigation shows that the demand to highly sensitive intelligent safety and defence system society in is huge, and to have only abroad be to count few company can produce the circumference safety-protection system that meets social demand in the whole world.
The wide-field full-optical fiber safety-protection system is mainly used in the safety monitoring monitoring of warehouse, airport, inflammable explosive article storehouse, dangerous cargo warehouse, national boundary, Aeronautics and Astronautics base, military base, industrial or agricultural base, bank, museum, prison, city tap-water, coal gas, rock gas, heat supply pipeline, oil, the long conveyance conduit of rock gas, electrical grid etc.The circumference safety guard all belongs to the application of broad domain all-optical fiber disturbance sensing and positioning system.
As the broad domain all-optical fiber disturbance sensing network system front end system; Full fiber reflection interferes distributed sensing system will wrap up in after the mixed signal of taking each noise like and disturbance information receives; Store in the data-signal buffer, can just become to estimate the good and bad key of this system performance to the processing that detected signal carries out promptly and accurately with identification this moment.
Summary of the invention
The purpose of the utility model is to overcome the deficiency that exists in the prior art; A kind of distributing optical fiber sensing application system self-adaptation type disturbing signal identification module is provided; It can be constructed the circumference security monitoring for important enforcement and zone signal identification accurately is provided, and carries out Security alert.
The technical scheme that provides according to the utility model; Said distributing optical fiber sensing application system self-adaptation type disturbing signal identification module; The signal buffer, data reader, signal denoising module and the denoising discrimination module that are positioned at front end are connected in series in proper order; " denying " output terminal of said denoising discrimination module links to each other with master controller; " being " output terminal of denoising discrimination module through endpoint detection module and end-point detection as a result discrimination module link to each other; End-point detection " denying " output terminal of discrimination module as a result links to each other with the input end of master controller; End-point detection " being " output terminal of discrimination module as a result links to each other with characteristic extracting module, said characteristic extracting module through characteristic contrast module and characteristic as a result discrimination module link to each other, said characteristic " deny " output terminal of discrimination module is as a result passed through category feature storehouse and characteristic and is contrasted module and link to each other; Characteristic " being " output terminal of discrimination module as a result links to each other with the pattern discrimination object module through the SVM identification module; " denying " output terminal of said pattern discrimination object module links to each other with characteristic extracting module, and " being " output terminal of pattern discrimination object module links to each other with the recognition result display-memory, and recognition result display-memory and master controller interconnect; Said master controller also interconnects with wavelet basis typelib, alarm and signal storage to be identified; The input end of denoising discrimination module directly also links to each other with the signal buffer, and master controller links to each other with the input end of endpoint detection module, and the output terminal of master controller links to each other with the signal denoising module.
Said master controller and signal denoising module and denoising discrimination module constitute small echo denoising self-adaptation ring; The signal that the denoising discrimination module is imported the signal buffer carries out the spectrum subtraction denoised signal with process signal denoising module and relatively judges; When if relatively judged result is not up to standard in the denoising discrimination module; Master controller changes the denoising method of signal denoising module or in the wavelet basis typelib, selects suitable wavelet basis; Through the denoising once more of signal denoising module, till denoising discrimination module comparison judged result is up to standard; Selectable wavelet basis comprises Daubechies wavelet basis and Symlets wavelet basis in the said wavelet basis typelib.
Said master controller and endpoint detection module and end-point detection discrimination module as a result constitute signal end detection self-adaptation ring; Strengthen through the garbage signal of endpoint detection module removal denoising discrimination module input and to characteristic signal; When end-point detection as a result discrimination module to the endpoint detection module input signal distinguishing when not up to standard; The parameter of master controller adjustment endpoint detection module is till end-point detection discrimination module detection as a result is up to standard.
Said characteristic contrast module, category feature storehouse and characteristic discrimination module as a result constitute signal identification self-adaptation ring; When characteristic when discrimination module decision signal characteristic does not exist in the category feature storehouse as a result, characteristic discrimination module as a result writes the category feature storehouse with new signal characteristic, otherwise will judge signal characteristic input SVM identification module.
Said characteristic extracting module, characteristic contrast module, SVM identification module and model results discrimination module constitute pattern-recognition self-adaptation ring; Said SVM identification module is sent recognition result into the model results discrimination module and is differentiated; When the model results discrimination module judges that SVM identification module recognition result does not meet; Then the model results discrimination module is through the feature extraction again of backout feature extraction module, so that SVM identification module recognition result and model results discrimination module are complementary.
Said characteristic extracting module is carried out feature extraction through wavelet character extraction method, two spectrum signature extraction method or Hilbert transform feature extraction method.
When recognition result display-memory during to master controller input hazard recognition signal, master controller is reported to the police through alarm, and shows recognition result through the recognition result display-memory.
The advantage of the utility model: interfere on the sensor-based system basis at distributed full fiber reflection; Mechanism and characteristics according to all kinds of disturbing signals in the environment; Denoising of employing spectrum subtraction or wavelet field denoising, the feature extraction of wavelet transformation energy threshold, artificial intelligence SVM comprehensive signal processing methods such as (support vector machine classifiers) have been proposed; Classification (harmful invasion or harmless disturbance) to signal is discerned, and identifies dangerous incident real-time and accurately; Rely on its strict discriminant function and a large amount of alternative disposal routes; Can under different environment, realize various acquired signal best processing and recognition effect; Promptly this method has very strong robustness and very high recognition correct rate; Can construct the circumference security monitoring for important enforcement and zone signal identification accurately is provided, and carry out Security alert.
Description of drawings
Fig. 1 is the utility model wide-field full-optical fiber safety-protection system disturbing signal adaptive model recognition device structured flowchart.
Fig. 2 is the utility model wide-field full-optical fiber safety-protection system disturbing signal adaptive model recognition device schematic diagram.
Fig. 3 is the scaling function and the wavelet function of db7 small echo.
Fig. 4 is the scaling function and the wavelet function of sym4 small echo.
Fig. 5 contains the climbing fence signal of noise and the climbing signal after the denoising.
Fig. 6 be contain noise cut fence protection network signal to cut fence protection network signal after the denoising.
Fig. 7 contains the heavy rain signal of noise and the heavy rain signal after the denoising.
Fig. 8 contains the punch press signal of noise and the punch press signal after the denoising.
Description of reference numerals: 100-system host; The 101-master controller; 201-signal buffer; The 202-data reader; 203-signal storage to be identified; 204-signal denoising module; 205-wavelet basis typelib; 206-denoising discrimination module; The 207-endpoint detection module; The 208-end-point detection is discrimination module as a result; The 209-characteristic extracting module; 210-characteristic contrast module; The 211-characteristic is discrimination module as a result; 212-category feature storehouse; The 213-SVM identification module; 214-model results discrimination module; 215-recognition result display-memory and 216-alarm.
Embodiment
Below in conjunction with concrete accompanying drawing and embodiment the utility model is described further.
As shown in Figure 1: the said signal buffer 201 that is positioned at front end is connected in series with data reader 202, signal denoising module 204 and denoising discrimination module 206 successively, and signal reader 202 reads behind the signal in the signal buffer 201 also with signal storage to signal storage 203 to be identified.Differentiate in order to let denoising discrimination module 206 carry out denoising, the output terminal of signal buffer 201 directly links to each other with denoising discrimination module 206." denying " output terminal of denoising discrimination module 206 links to each other with master controller 101; " being " output terminal of denoising discrimination module 206 links to each other with endpoint detection module 207; The output terminal of master controller 101 links to each other with signal denoising module 204; When the denoising of denoising discrimination module 206 decision signal denoising modules 204 was not up to standard, the denoising mode of master controller 101 conditioning signal denoising modules 204 was till the 206 judgment signal denoisings of denoising discrimination module are up to standard.207 pairs of signals of endpoint detection module are further handled; Can remove useless point strengthens characteristic signal simultaneously; Endpoint detection module 207 and end-point detection discrimination module 208 as a result link to each other; End-point detection " denying " output terminal of discrimination module 208 as a result links to each other with master controller 101; End-point detection " being " output terminal of discrimination module 208 as a result links to each other with characteristic extracting module 209, and master controller 101 links to each other with endpoint detection module 207, thus when end-point detection as a result discrimination module 208 detect endpoint detection module 207 results when not up to standard; Return endpoint detection module 207 through master controller 101 and carry out end-point detection once more, till end-point detection discrimination module 208 differentiations as a result are up to standard.The output terminal of characteristic extracting module 209 links to each other with characteristic contrast module 210; Carry out feature extraction through characteristic extracting module 209; Characteristic contrast module 210 is touched 212 with characteristic result differentiation and is linked to each other; Said characteristic " denying " output terminal of discrimination module 212 as a result links to each other with category feature storehouse 211, and category feature storehouse 211 links to each other with characteristic contrast module 210, and characteristic " being " output terminal of discrimination module 212 as a result links to each other with SVM identification module 213; SVM identification module 213 links to each other with model results discrimination module 214; " denying " output terminal of said model results discrimination module 214 links to each other with characteristic extracting module 209, and " being " output terminal of model results discrimination module 214 links to each other with recognition result display-memory 215, and said recognition result display-memory 215 links to each other with master controller 101.Master controller 101 interconnects with wavelet basis typelib 205, alarm 216 and signal storage to be identified 203.
Data reader 202, will be imported data and deposit to signal storage 203 to be identified when it delivers to signal denoising module 204 with data from signal buffer 201 reading of data, so that carry out corresponding subsequent analysis when finding danger signal later on.Signal storage 203 to be identified links to each other with master controller 101 simultaneously, shows its corresponding recognition result by master controller 101 controls.
Said master controller 101 constitutes small echo denoising self-adaptation ring with signal denoising module 204 and denoising discrimination module 206; The signal that the signal of 206 pairs of signal buffers of denoising discrimination module, 201 inputs and process signal denoising module 204 are carried out after the spectrum subtraction denoising is relatively judged; When if relatively judged result is not up to standard in the denoising discrimination module 206; Promptly not obvious through the spectrum subtraction denoising effect of signal denoising module 204; Master controller 101 changes the denoising method of signal denoising modules 204 or in wavelet basis typelib 205, selects suitable wavelet basis at this moment; Through signal denoising module 204 denoising once more under small echo denoising mode, till denoising discrimination module 206 comparison judged results are up to standard; Selectable wavelet basis comprises Daubechies wavelet basis and Symlets wavelet basis in the said wavelet basis typelib 205.The setting judgment threshold is provided with according to the environment of input signal in the denoising discrimination module 206.
Said master controller 101 detects the self-adaptation ring in discrimination module 208 formation signal ends as a result with endpoint detection module 207 and end-point detection; Strengthen through the garbage signal of endpoint detection module 207 removal denoising discrimination modules 206 inputs and to characteristic signal; When end-point detection when 208 pairs of endpoint detection module 207 input signal distinguishings of discrimination module are not up to standard as a result; The parameter of master controller 101 adjustment endpoint detection modules 207 is till end-point detection discrimination module 207 detections as a result are up to standard.Renting of endpoint detection module 207 is the further processing after the signal denoising, mainly is garbage signal (comprising some singular points) and the characteristic reinforcement of removing in the signal of useful signal; Amplitude and continuum length according to signal are removed garbage signal and are kept useful signal.The parameter of master controller 101 adjustment endpoint detection modules 207 comprises continuum length, and end-point detection is used for the threshold value that target judges as a result in the discrimination module 208 and is provided with according to site environment.
Said characteristic contrast module 210, category feature storehouse 211 and characteristic discrimination module 212 as a result constitute signal identification self-adaptation ring; When characteristic when discrimination module 212 decision signal characteristics do not exist in category feature storehouse 211 as a result, characteristic discrimination module 212 as a result writes category feature storehouse 211 with new signal characteristic, otherwise will judge signal characteristic input SVM identification module 213.SVM identification module 213 calculates judgement with the eigenwert in eigenwert of being extracted and the category feature storehouse 211 through the SVM principle, and then provides recognition result.
Said characteristic extracting module 209, characteristic contrast module 210, SVM identification module 213 and model results discrimination module 214 constitute pattern-recognition self-adaptation rings; Said SVM identification module 213 is sent recognition result into model results discrimination module 214 and is differentiated; When model results discrimination module 214 judges that SVM identification module 213 recognition results do not meet; Then model results discrimination module 214 is through backout feature extraction module 209 feature extraction again, so that SVM identification module 213 recognition results and model results discrimination module 214 are complementary.Said characteristic extracting module 209 is carried out feature extraction through wavelet character extraction method, two spectrum signature extraction method or Hilbert transform feature extraction method; When characteristic extracting module 209 adopts the different character method for distilling, can access corresponding characteristic quantity.When said characteristic extracting module 209 was carried out feature extraction through wavelet character extraction method, two spectrum signature extraction method or Hilbert transform feature extraction method, the characteristic quantity that obtains was respectively energy feature, Hilbert spectrum signature and two spectrum signature of WAVELET PACKET DECOMPOSITION.
Said master controller 101 is in charge of the operation of total system, comprises optical signal transmitting and the realization that receives control and disturbance location; Recognition result according to artificial intelligence SVM identification module 213 carries out the Different control operation; For example when the analysis result to the environment input signal is shown as harmless disturbing signal; Master controller 101 does not just carry out the corresponding subsequent control operation, and continues work as usual; If the analysis result to the environment measuring signal is shown as danger signal, master controller 101 just starts alarm 216 so, controls signal storage 203 to be identified simultaneously, and recognition result is shown through recognition result display-memory 215.
As shown in Figure 2: the master controller 101 in the utility model, data reader 202, signal storage to be identified 203, signal denoising module 204, endpoint detection module 207, characteristic extracting module 209, SVM identification module 210, recognition result display-memory 215 are included in the system host 100.The input of data-signal buffer 201 one tunnel output welding system main frames 100, system host 100 has two output signals, and one is output as recognition result display-memory 215, and another exports device 216 taking alarm.It is available that the hardware of most of module all has general market product in the native system.
The utility model is interfered on the sensor-based system basis at distributed full fiber reflection; Mechanism and characteristics according to all kinds of disturbing signals in the environment; Denoising of employing spectrum subtraction or wavelet field denoising, the feature extraction of wavelet transformation energy threshold, artificial intelligence SVM comprehensive signal processing methods such as (support vector machine classifiers) have been proposed; Classification (harmful invasion or harmless disturbance) to signal is discerned, and identifies dangerous incident real-time and accurately.The utility model implementation procedure comprises the content of two aspects; 1, from signal buffer 201, the mixed signal of carrying each noise like and disturbance information is read out, carry out signal denoising through signal denoising module 204; Carrying out end-point detection through endpoint detection module 207 handles; Carry out the extraction of eigenwert through characteristic extracting module 209 then, get into SVM identification module 213 at last, identify the type and the character of input signal; 2, when signal is handled and discerned, deposit it in signal storage 203 to be identified, after the recognition result analysis of treating input signal finishes, whether this input signal is discerned classification by master controller 101 decisions and show.
Specific embodiment:
The embodiment of the utility model is described and the utility model is done further explanation with this example.This example is an experimental prototype.System host comprises digital signal software pre-processing module (in the signal denoising process involved all modules), characteristic extracting module, artificial intelligence recognition system.This example uses a computer the machine model for associating ThinkCentre M6300s; Computer monitor: 22 inches; Computer CPU: Intel Duo i3 2100, operating system: Windows 7 professional versions, memory size: 2GB; Video card core: NVIDIA GeForce G405, hard-disk capacity: 320GB.Specifically constitute: 201 one tunnel outputs of data-signal buffer connect the input of this main frame; Read and store for the input data; Carry out the spectrum subtraction denoising then and also can select the small echo denoising; Can use acquiescence Daubechies wavelet basis (Fig. 3) to carry out small echo denoising (undesirable like denoising effect, as can to select Symlets small echo (Fig. 4) etc. for use) earlier.Denoising result is carried out carrying out WAVELET PACKET DECOMPOSITION feature extraction wavelet basis after the end-point detection through endpoint detection module 207 can be selected; Can select Daubechies wavelet basis, Symlets small echo (Fig. 4) etc. for use) with the eigenwert of extracting; Through sending into SVM identification module 213 after the characteristic contrast module 210; Said SVM identification module 213 is made up of NBBT-SVM (non-equilibrium binary-tree support vector machine) sorter; Recognition result display-memory 215 can return recognition result to master controller 101; If recognition result is then master controller 101 startup alarms 216 of danger signal,, wait for the arrival of identification requirement next time if non-danger (harmless) signal then continues operate as normal.Fig. 5, Fig. 6 are respectively the noisy and denoising waveform of danger signal (climb the fence signal and cut the net signal).Fig. 7, Fig. 8 are respectively the noisy and denoising waveform of non-danger signal (disturbing signal during the large group's punch press work forever of heavy rain signal and Wuxi).After unlike signal through Fig. 5 ~ Fig. 8 carries out denoising, end-point detection, feature extraction and pattern-recognition; Can accurately discern the type of input signal; Master controller 101 can carry out alarm through alarm 216, and can show the output recognition result through recognition result display register 215.
Therefore; The utility model spectrum of use is subtracted each other signal analysis, the Pattern recognition principle of means such as the signal processing method comprehensive utilization of method that denoising, WAVELET PACKET DECOMPOSITION energy feature extract, artificial intelligence NBBT-SVM (non-equilibrium binary-tree support vector machine); Can effective monitoring major facility and zone; Identify critical facility and the dangerous situation that the zone occurs timely and accurately on every side, be with a wide range of applications.
Claims (7)
1. distributing optical fiber sensing application system self-adaptation type disturbing signal identification module; It is characterized in that: the signal buffer (201), data reader (202), signal denoising module (204) and denoising discrimination module (206) order that are positioned at front end are connected in series; " denying " output terminal of said denoising discrimination module (206) links to each other with master controller (101); " being " output terminal of denoising discrimination module (206) through endpoint detection module (207) and end-point detection as a result discrimination module (208) link to each other; End-point detection " denying " output terminal of discrimination module (208) as a result links to each other with the input end of master controller (101); End-point detection " being " output terminal of discrimination module (208) as a result links to each other with characteristic extracting module (209); Said characteristic extracting module (209) through characteristic contrast module (210) and characteristic as a result discrimination module (212) link to each other; Said characteristic " denying " output terminal of discrimination module (212) as a result links to each other with characteristic contrast module (210) through category feature storehouse (211); Characteristic " being " output terminal of discrimination module (212) as a result links to each other with pattern discrimination object module (214) through SVM identification module (213); " denying " output terminal of said pattern discrimination object module (214) links to each other with characteristic extracting module (209), and " being " output terminal of pattern discrimination object module (214) links to each other with recognition result display-memory (215), and recognition result display-memory (215) interconnects with master controller (101); Said master controller (101) also interconnects with wavelet basis typelib (205), alarm (216) and signal storage to be identified (203); The input end of denoising discrimination module (206) directly also links to each other with signal buffer (201), and master controller (101) links to each other with the input end of endpoint detection module (207), and the output terminal of master controller (101) links to each other with signal denoising module (204).
2. distributing optical fiber sensing application system self-adaptation type disturbing signal identification module according to claim 1 is characterized in that: said master controller (101) constitutes small echo denoising self-adaptation ring with signal denoising module (204) and denoising discrimination module (206); The signal that denoising discrimination module (206) is imported signal buffer (201) carries out the spectrum subtraction denoised signal with process signal denoising module (204) and relatively judges; When if relatively judged result is not up to standard in the denoising discrimination module (206); Master controller (101) changes the denoising method of signal denoising module (204) or in wavelet basis typelib (205), selects suitable wavelet basis; Through signal denoising module (204) denoising once more, till denoising discrimination module (206) comparison judged result is up to standard; Selectable wavelet basis comprises Daubechies wavelet basis and Symlets wavelet basis in the said wavelet basis typelib (205).
3. distributing optical fiber sensing application system self-adaptation type disturbing signal identification module according to claim 1 is characterized in that: said master controller (101) detects the self-adaptation ring in discrimination module (208) formation signal end as a result with endpoint detection module (207) and end-point detection; Strengthen through the garbage signal of endpoint detection module (207) removal denoising discrimination module (206) input and to characteristic signal; When end-point detection as a result discrimination module (208) to endpoint detection module (207) input signal distinguishing when not up to standard; The parameter of master controller (101) adjustment endpoint detection module (207) is till end-point detection discrimination module (207) detection as a result is up to standard.
4. distributing optical fiber sensing application system self-adaptation type disturbing signal identification module according to claim 1 is characterized in that: said characteristic contrast module (210), category feature storehouse (211) and characteristic discrimination module (212) as a result constitute signal identification self-adaptation ring; When characteristic as a result discrimination module (212) decision signal characteristic in category feature storehouse (211) when not existing; Characteristic discrimination module (212) as a result writes category feature storehouse (211) with new signal characteristic, otherwise will judge signal characteristic input SVM identification module (213).
5. distributing optical fiber sensing application system self-adaptation type disturbing signal identification module according to claim 1 is characterized in that: said characteristic extracting module (209), characteristic contrast module (210), SVM identification module (213) and model results discrimination module (214) constitute pattern-recognition self-adaptation ring; Said SVM identification module (213) is sent recognition result into model results discrimination module (214) and is differentiated; When model results discrimination module (214) judges that SVM identification module (213) is not when recognition result meets; Then model results discrimination module (214) is through backout feature extraction module (209) feature extraction again, so that SVM identification module (213) recognition result and model results discrimination module (214) are complementary.
6. according to claim 1 or 5 described distributing optical fiber sensing application system self-adaptation type disturbing signal identification modules, it is characterized in that: said characteristic extracting module (209) is carried out feature extraction through wavelet character extraction method, two spectrum signature extraction method or Hilbert transform feature extraction method.
7. distributing optical fiber sensing application system self-adaptation type disturbing signal identification module according to claim 1; It is characterized in that: when recognition result display-memory (215) during to master controller (101) input hazard recognition signal; Master controller (101) is reported to the police through alarm (216), and shows recognition result through recognition result display-memory (215).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011203607287U CN202404694U (en) | 2011-09-23 | 2011-09-23 | Adaptive disturbance signal identification module of distributing type optical fiber sensing application system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011203607287U CN202404694U (en) | 2011-09-23 | 2011-09-23 | Adaptive disturbance signal identification module of distributing type optical fiber sensing application system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN202404694U true CN202404694U (en) | 2012-08-29 |
Family
ID=46702329
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011203607287U Withdrawn - After Issue CN202404694U (en) | 2011-09-23 | 2011-09-23 | Adaptive disturbance signal identification module of distributing type optical fiber sensing application system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN202404694U (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102509403A (en) * | 2011-09-23 | 2012-06-20 | 无锡科晟光子科技有限公司 | Device for recognizing disturbing signal of wide-area all-fiber security system in self-adapting mode |
| CN104240455A (en) * | 2014-08-07 | 2014-12-24 | 北京航天控制仪器研究所 | Method for identifying disturbance event in distributed type optical fiber pipeline security early-warning system |
| CN105931402A (en) * | 2016-06-27 | 2016-09-07 | 上海波汇科技股份有限公司 | Optical fiber perimeter intrusion monitoring method based on image recognition |
-
2011
- 2011-09-23 CN CN2011203607287U patent/CN202404694U/en not_active Withdrawn - After Issue
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102509403A (en) * | 2011-09-23 | 2012-06-20 | 无锡科晟光子科技有限公司 | Device for recognizing disturbing signal of wide-area all-fiber security system in self-adapting mode |
| CN102509403B (en) * | 2011-09-23 | 2013-07-31 | 无锡科晟光子科技有限公司 | Device for recognizing disturbing signal of wide-area all-fiber security system in self-adapting mode |
| CN104240455A (en) * | 2014-08-07 | 2014-12-24 | 北京航天控制仪器研究所 | Method for identifying disturbance event in distributed type optical fiber pipeline security early-warning system |
| CN105931402A (en) * | 2016-06-27 | 2016-09-07 | 上海波汇科技股份有限公司 | Optical fiber perimeter intrusion monitoring method based on image recognition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110390357A (en) | A kind of DTU safety monitoring method based on side channel | |
| CN104965235A (en) | Security check system and method | |
| CN116781430B (en) | Network information security system and method for gas pipe network | |
| CN103235942A (en) | Facial recognition method applied to entrance guard | |
| CN103236127A (en) | Fiber fence intrusion monitoring system and pattern recognition method thereof | |
| CN113283344A (en) | Mining conveying belt deviation detection method based on semantic segmentation network | |
| CN205788364U (en) | A kind of Smart Home security alarm gate control system | |
| CN103279999A (en) | Card reading system based on face recognition | |
| CN110796372A (en) | Industrial enterprise safety production risk identification system and method thereof | |
| CN108009488A (en) | The street security joint defense system of Behavior-based control analysis | |
| CN107256357A (en) | The detection of Android malicious application based on deep learning and analysis method | |
| CN106023372A (en) | Library access management system based on face recognition | |
| CN202404694U (en) | Adaptive disturbance signal identification module of distributing type optical fiber sensing application system | |
| CN107764318A (en) | Method for detecting abnormality and Related product | |
| CN106910152A (en) | A kind of bank's security a suspect's intelligent identifying system based on WIFI detections | |
| CN112101234A (en) | Detection code matching processing method and image code joint detection system | |
| CN201679381U (en) | Safety box based on face recognition | |
| Surantha et al. | An IoT based house intruder detection and alert system using histogram of oriented gradients | |
| CN107220623A (en) | A kind of face identification method and system | |
| CN203287954U (en) | Card punching system based on human face recognition | |
| CN103218863A (en) | Pattern recognition based barrier-free channel machine bidirectional detection method | |
| Iqbal et al. | Artificial intelligence solutions to detect fraud in healthcare settings: a scoping review | |
| CN116389148B (en) | Network security situation prediction system based on artificial intelligence | |
| CN102509403B (en) | Device for recognizing disturbing signal of wide-area all-fiber security system in self-adapting mode | |
| CN103825875A (en) | Virtual machine detection method for vaccine inoculation strategy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| AV01 | Patent right actively abandoned |
Granted publication date: 20120829 Effective date of abandoning: 20130731 |
|
| RGAV | Abandon patent right to avoid regrant |