CN202650186U - Forest fire early warning system based on uncooled infrared temperature detector - Google Patents
Forest fire early warning system based on uncooled infrared temperature detector Download PDFInfo
- Publication number
- CN202650186U CN202650186U CN 201220069019 CN201220069019U CN202650186U CN 202650186 U CN202650186 U CN 202650186U CN 201220069019 CN201220069019 CN 201220069019 CN 201220069019 U CN201220069019 U CN 201220069019U CN 202650186 U CN202650186 U CN 202650186U
- Authority
- CN
- China
- Prior art keywords
- uncooled
- temperature
- hygrosensor
- early warning
- alarm server
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Alarm Systems (AREA)
Abstract
The utility model provides a forest fire early warning system based on an uncooled infrared temperature detector. The forest fire early warning system comprises at least one uncooled infrared temperature detector, a network and an alarm server. The uncooled infrared temperature detector and the alarm server are connected with the network through respective network transmission equipment. The forest fire early warning system of the utility model can solve a problem of difficult determination of an alarm threshold caused by the uncertainty of a detection distance, atmospheric transmittance and air humidity, and the rate of false alarm and missing alarm is effectively reduced.
Description
Technical field
The utility model relates to a kind of forest fires early warning system, particularly a kind of forest fires early warning system based on the uncooled ir hygrosensor.
Background technology
China is the lower but country of many forest fires of afforestation rate.In recent years, along with the reinforcement of environmental protection consciousness, conservation of resources consciousness and the progress of science and technology, take put out a fire to save life and property as main passive fire protection to changing take early warning as main active fire prevention.Particularly under the drive of the technology such as infotech, Aero-Space, China's forest fire protection in monitoring, warning aspect obtained significant progress, from manually to put prevention first to set up unattended robotization early warning system as the main forest fire protection basic principle that changes and determined " beat long ago, little, beaten ".
Because the condition of a fire is subjected to the impact of the conditions such as Protean weather, meteorology, vegetation, geographical environment very large, the reason that causes the condition of a fire is extremely complicated and uncertain abrim, so how to find to lie concealed the various infrared sources that in forest, might cause forest fire early, it is the difficult problem in face of putting in the whole world.Mainly exist in the world at present following several forest fire protection monitoring mode:
1 artificial loud and clear prestige
Monitor the condition of a fire by the artificial loud and clear prestige whistle that places the mountain top, this is a kind of more traditional mode.In order to ensure that the artificial loud and clear prestige whistle has the good visual field, whistle tower all can be based upon remote, thickly forested mountains and the mountain top of the rare mark of the signs of human habitation usually, and the duty personnel generally is faced with the difficult problems such as clothing, food, lodging and transportion--basic necessities of life, children's education.At present, the artificial loud and clear prestige whistle of China is fewer and feweri.
2 cruise in the air
Along with the continuous enhancing of China's economic technology strength, in recent years, the mode of China by cruising in the air carried out air-borne early warning to the forest zone that has possessed high forest fire danger class.But the climate that cruises, wait factor affecting larger round the clock in the air, the cost costliness of in addition cruising so its monitored density is limited, is carried out 7 * 24 hours daily monitoring important task to forest fires in the large space scope that is difficult to bear the responsibility.
3 satellite monitorings
China has generally adopted satellite monitoring in some large forest zones at present.In macroscopic aspect, the satellite Forest Fire Monitoring is to assess after forest fire appealing trend analysis, the forest fires calamity etc., and very effective technological means is provided.But because satellite is away from earth surface and condition of a fire generation area, its resolution to temperature is common only about 1000 square meters, very difficult discovery might cause hidden fire and the little fire of forest fire, is difficult to satisfy the basic demand of forest fire protection " long ago, beat little, beaten ".
The 4 forest fires early warning based on warming, cigarette propagated sensation sensor
The condition of a fire is identified and found to warming, cigarette propagated sensation sensor by the concentration of the gases such as near the temperature the acquisition sensor, dust concentration, carbon monoxide and carbon dioxide.This mode is more effective for the fire-fighting of city, building, fixed-site etc., but is not suitable for the deployment of carrying out magnanimity in the large space scope, the viability under its field extreme environment, life cycle etc., also is difficult to satisfy the requirement of forest fire protection.
The 5 video monitorings based on visible light
In recent years, the video frequency monitoring system based on visible light has all been installed in many forest zones by China, remedies according to this deficiencies of monitoring means in the forest fires early warning such as the artificial lookout tower whistle, satellite.Forest fires video monitoring means based on visible light mainly is according to digital image processing techniques, and the morphological feature of fire image is compared analysis.These means have overcome the defective that the other technologies means exist in some aspects to a certain extent, but owing to not possessing temp sensing function based on the video frequency monitoring system of visible light, and the cloud and mist that rises layer by layer in the mountain forest is very similar to the morphological feature of forest fires, the randomness of cloud and mist rise or drift is very strong again, is difficult to be got rid of by default intellectual analysis rule.So the rate of false alarm of this system and rate of failing to report are usually all higher.In actual applications, also need to be equipped with a large amount of duty personnel and ceaselessly watch monitoring image, do not reach the technical requirement of forest fires automatic early-warning system.
6 uncooled ir hygrosensors (in the utility model also can referred to as infrared temperature sensor)
Because above-mentioned various technological means all exist insoluble problem in various a period of time more or less in forest fires early detection and early warning, so more domestic forest zones begin to attempt to realize large space scope interior early detection and early warning to forest fires with the uncooled ir hygrosensor.By the physics blackbody radiation theory as can be known, any temperature is higher than the object of absolute zero (273.15 ℃), and all infrared energy is sent in the space towards periphery, and the value of infrared energy and the surface temperature of object are closely related.Infrared temperature sensor automatically carries out real-time measurements continual, 7 * 24 hours just according to this physics principle to the body surface temperature of monitoring.But in actual applications, this technological means still exists following problem:
1) because the infrared energy that sends of object can be subject to the impact of the factor that distance, air humidity, atmospheric transmissivity etc. can't determine in advance when arriving infrared temperature sensor, and produces uncertain energy loss.So in the large space monitoring range (0-5 kilometer), be difficult to the alarm temperature under definite different distance place, the different humidity condition.If alarm temperature arranges lowlyer, rate of false alarm will increase; If alarm temperature arranges highlyer, rate of failing to report will increase.Especially, the temperature of the open-air naked light far away apart from infrared temperature sensor may be lower than the temperature apart from infrared temperature sensor stone nearer, that certain piece has been drying heat, thereby has not only produced wrong report (stone close together, that be drying heat), but also produced and fail to report (distant open-air naked light).
2) the uncooled ir hygrosensor can not identify that heat radiation stems from open-air naked light or such as non-forest fires (open-air naked light) class heat radiations such as car engine and tire, kitchen smokes, thereby has caused the increase of rate of false alarm (misrepresenting deliberately rate).
The utility model content
The utility model is a kind of forest fires early warning system based on the uncooled ir hygrosensor, the problem that is difficult to determine alarm threshold value that the uncertainty owing to detection range, atmospheric transmissivity, atmospheric humidity causes be can solve, thereby rate of false alarm and rate of failing to report effectively reduced.
A kind of forest fires early warning system based on the uncooled ir hygrosensor, comprise at least one uncooled ir hygrosensor, network, Alarm Server, described uncooled ir hygrosensor and described Alarm Server pass through respectively network communication and transmission equipment and described network connection separately.
Described Alarm Server can comprise the data acquisition module that links to each other successively, characteristic extracting module and alert service module, described data acquisition module is by the temperature data in the detective field of view angle of described network collection uncooled ir hygrosensor and be transferred to characteristic extracting module, described characteristic extracting module is extracted the temperature fluctuation feature in the detective field of view angle of uncooled ir hygrosensor and is judged whether according to this to have occured the unusual fluctuations of temperature from described temperature data, if the temperature fluctuation feature in the described detective field of view angle shows as temperature jump, then show the unusual fluctuations that temperature has occured, if it is continually varying that the temperature fluctuation feature in the described detective field of view angle shows as temperature, then show the unusual fluctuations that do not have occurrence temperature; Described characteristic extracting module passes to the alert service module with the unusual fluctuations situation of temperature, in time sends warning by the alert service module.
Described characteristic extracting module can be extracted the temperature fluctuation feature from the temperature data in the detective field of view angle of uncooled ir hygrosensor acquisition by Sobel Operator.
Described Alarm Server can also comprise characteristics analysis module, and described characteristic extracting module is connected with described alert service module by described characteristics analysis module; Described characteristics analysis module judges whether the current temperature fluctuation feature of surveying the moment and a upper detection moment has similarity, if two temperature fluctuation feature dissmilarities of surveying constantly, then continue to judge whether the dissmilarity of described temperature fluctuation feature is derived from the class heat radiation, if not be derived from the class heat radiation, then judged result is sent to the alert service module.
Above-mentioned forest fires early warning system, if two temperature fluctuation feature dissmilarities of surveying constantly, described continuation judges whether the dissmilarity of described temperature fluctuation feature is derived from the class heat radiation, can be to compare by the temperature fluctuation feature that will extract and the sample characteristics in the sample fluctuation feature database to realize; Described sample fluctuation feature database is the database that is connected with described characteristics analysis module;
And/or described sample fluctuation feature database can have ability of self-teaching.
Above-mentioned forest fires early warning system, if two temperature fluctuation feature dissmilarities of surveying constantly, described continuation judges whether the dissmilarity of described temperature fluctuation feature is derived from the class heat radiation, can be by with the temperature fluctuation feature of extracting recently realization relative to the temperature fluctuation feature of previous moment same space coordinate.
Above-mentioned forest fires early warning system can also comprise digital tripod head, and each described uncooled ir hygrosensor is connected with digital tripod head.
Above-mentioned forest fires early warning system, can also comprise the warning client, described Alarm Server is connected by Wide Area Network with the warning client, and provides alert service to the warning client, and described warning client is forest fire protection alarm and command systems at different levels and/or third party system;
And/or the alert service mode that described Alarm Server provides to the warning client can be the SDK mode, or message subscribing manner, or in the message push mode on one's own initiative to warning client push warning message.
Above-mentioned forest fires early warning system also comprises encoder, and described scrambler is connected with the uncooled ir hygrosensor, and described demoder is connected with Alarm Server and/or warning client.
A kind of forest fires early warning system based on the uncooled ir hygrosensor, can comprise the bottom Alarm Server and with at least one deck upper strata Alarm Server of described bottom Alarm Server by network connection, described bottom Alarm Server is the Alarm Server of above-mentioned forest fires early warning system.
Technique effect of the present utility model
A kind of forest fires early warning system based on the uncooled ir hygrosensor of the present utility model, because its Alarm Server can not rely on detection range, atmospheric transmissivity, atmospheric humidity is judged the temperature anomaly fluctuation, therefore can solve because detection range, atmospheric transmissivity, the uncertainty of atmospheric humidity and the problem that is difficult to determine alarm threshold value that causes, effectively solved the difficult problem that the threshold formula warning system based on physics (temperature) dimension faces in actual applications, thereby realized in the large space scope, under non-definite distance and the meteorological condition, round-the-clock monitoring and early warning to forest fires have reduced rate of false alarm and rate of failing to report effectively.
The Alarm Server of further program of the utility model is by data acquisition module, these hardware of characteristic extracting module and alert service module consist of, can not rely on detection range, atmospheric transmissivity, atmospheric humidity is judged the temperature anomaly fluctuation, therefore can solve because detection range, atmospheric transmissivity, the uncertainty of atmospheric humidity and the problem that is difficult to determine alarm threshold value that causes, effectively solved the difficult problem that the threshold formula warning system based on physics (temperature) dimension faces in actual applications, thereby realized in the large space scope, under non-definite distance and the meteorological condition, round-the-clock monitoring and early warning to forest fires have reduced rate of false alarm and rate of failing to report effectively.
Further program of the utility model adopts Sobel Operator to extract the temperature fluctuation feature from temperature data, owing to can describe the profile of high-temperature area by Sobel Operator, therefore can be used for the temperature jump in the identification detective field of view angle, and the information that whether spreads of high-temperature region.
The Alarm Server of further program of the utility model passes through characteristics analysis module, judge whether the current temperature fluctuation feature of surveying the moment and a upper detection moment has similarity, can correctly identify the class heat radiation of non-forest fires (non-open-air naked light), effectively get rid of the wrong report of uncooled ir hygrosensor for non-forest fires (open-air naked light) class heat radiations (such as the kitchen smoke in the non-forest zone, car engine and tire, ice and snow etc.), greatly improved practicality, advance based on the forest fires early warning system of uncooled ir hygrosensor.
Further program of the utility model adopts sample characteristics and the actual temperature fluctuation feature that detects of storing in the sample fluctuation feature database to compare, whether the position that can determine the occurrence temperature unusual fluctuations is in the forest zone, if be not in the forest zone, then naturally got rid of the possibility that forest fires occur, if be in the forest zone, forest fires have then very likely really occured, therefore send warning message, can get rid of more accurately the class heat radiation by this sample fluctuation feature database, thereby reduce the rate of misrepresenting deliberately.
The sample fluctuation feature database of further program of the utility model has ability of self-teaching, do not need the professional to carry out a large amount of operations, can automatically increase fluctuation characteristic amount wherein, saved a large amount of manpower and materials, enrich the sample fluctuation feature database, improved efficient.
Further program of the utility model compares the temperature fluctuation feature of extraction and the temperature fluctuation feature of previous moment same space coordinate, thereby can judge the variation of condition of a fire generation territorial scope, can monitor out just thus whether the intensity of a fire has occured to spread and the concrete condition of fire spreading.
The digital tripod head of further program of the utility model can assist the uncooled ir hygrosensor that its peripheral region is carried out full monitoring within the vision, has farthest utilized the uncooled ir hygrosensor.
Further program of the utility model is connected with the warning client Alarm Server by Wide Area Network, made things convenient for all departments that need to obtain forest zone condition of a fire information can obtain in real time, easily forest zone condition of a fire information, when the condition of a fire occurs, can take measures the very first time; Alarm Server has the ability that interconnects between good system, can provide alert service to the third party system in many ways.Further program of the utility model is connected with the warning client Alarm Server in several ways by Wide Area Network, the convenient connection, and can the fast transport real-time information.
The scrambler of further program of the utility model can directly become the infrared imaging data compression suit to send to Alarm Server or warning client (comprising third party system etc.) again behind the data layout that Wide Area Network transmits, thereby can easily directly with infrared image form displays image information on Alarm Server and/or warning client, make things convenient for monitoring; Further program of the utility model, described digital tripod head can also pass through scrambler connecting wan network, has reduced the equipment of the direct connecting wan network of digital tripod head, reduced assembly, reduce cost, and reduced the complicacy of equipment, thereby reduced rate of breakdown.
A kind of forest fires early warning system of the present utility model, a plurality of Alarm Servers can be carried out classification according to management area and/or supervisory level disposes, can avoid because the warning message that a variety of causes causes is out in the cold or delay, thereby can to the full extent with in the shortest time warning message be sent to each competent authorities.
Description of drawings
Fig. 1 a is the block diagram of an embodiment of forest fires early warning system of the present utility model.
Fig. 1 b is the block diagram of an embodiment of forest fires early warning system of the present utility model.
Fig. 2 is the workflow diagram of the Alarm Server of an embodiment of the present utility model.
Embodiment
The utility model is a kind of forest fires early warning system based on the uncooled ir hygrosensor, in the large space scope, be used for the deficiency that the forest fires early detection exists for the uncooled ir hygrosensor, provided one passed through experimental verification, can reduce significantly the technical scheme of forest fires rate of false alarm and rate of failing to report, can solve because detection range, atmospheric transmissivity, the uncertainty of atmospheric humidity and the problem that is difficult to determine alarm threshold value that causes, thereby rate of false alarm and rate of failing to report have effectively been reduced, and can correctly identify the class heat radiation of non-forest fires, reduce the rate of misrepresenting deliberately.
The utility model is done following elaboration to some nouns and the term that relate in the utility model before the technical solution of the utility model is summarized:
A, uncooled ir hygrosensor: the uncooled ir hygrosensor is to utilize the thermoradiation efficiency of testee (any temperature is higher than the object of absolute zero) and the energy-probe made.The uncooled ir hygrosensor is operated in the 8-14 mu m waveband, on this service band, and the radiation energy of detector integrates incident, and radiation energy is converted into the electric signal that is directly proportional with emittance.Can convert the electrical signal to surface temperature into testee by heat-conduction equation.The uncooled ir hygrosensor can by un-cooled infrared focal plane array (also claiming matrix), be realized infrared imaging.
B, frame: frame is the thermal-induced imagery that a width of cloth is comprised of the pixel in the uncooled ir hygrosensor field angle (such as 320 * 240 pixels), and thermal-induced imagery has reflected the monitoring visual field of uncooled ir hygrosensor.
C, the condition of a fire: the condition of a fire in the utility model refers to that open-air naked light (being forest fires) maybe can cause the burning things which may cause a fire disaster of open-air naked light, such as high temperature smog etc.
D, class heat radiation: class heat radiation of the present utility model refers to non-open-air naked light, neither cause other heat radiation of the burning things which may cause a fire disaster of open-air naked light, reflective such as the ice face, the heat of friction on tire friction highway ground, the car engine heating shines upon the heat radiation of stone accumulation etc.
Below in conjunction with accompanying drawing embodiment of the present utility model is described in detail.
Fig. 1 a is the block diagram of an embodiment of forest fires early warning system of the present utility model, and real line represents physical line wherein virtually connects the transmission of wiring representative information.A kind of forest fires early warning system based on the uncooled ir hygrosensor among Fig. 1 a comprises at least one uncooled ir hygrosensor, network (LAN, or the wide-area communication transmission network also is called for short Wide Area Network), Alarm Server.Described uncooled ir hygrosensor and described Alarm Server pass through respectively network communication and transmission equipment and described network connection separately.Described Alarm Server comprises the data acquisition module that links to each other successively, characteristic extracting module and alert service module, described data acquisition module is by the temperature fluctuation feature in the detective field of view angle of described network collection uncooled ir hygrosensor and be transferred to characteristic extracting module, described characteristic extracting module is extracted temperature fluctuation feature in the detective field of view angle of uncooled ir hygrosensor (in the utility model also can referred to as fluctuation characteristic or feature) and is judged whether according to this to have occured the unusual fluctuations of temperature from described temperature data, if the temperature fluctuation feature in the described detective field of view angle shows as temperature jump (temperature is discontinuous), then show the unusual fluctuations that temperature has occured, if it is that continuously smooth changes that the temperature fluctuation feature in the described detective field of view angle shows as temperature, namely do not have temperature jump, then show the unusual fluctuations that do not have occurrence temperature.Particularly, if not there is not the temperature anomaly fluctuation in the detection viewing field angular region of refrigeration infrared temperature sensor, so in the data of its acquisition, it is continuously smooth that temperature fluctuation feature in the field angle scope shows as temperature variation, not sudden change, (temperature such as the river is usually lower even the temperature of zones of different is variant, and the temperature on road surface is because Exposure to Sunlight is usually higher, therefore often the variation of temperature is larger at the intersection on river and road surface), but at the intersection of zones of different, its temperature variation still is continuous; If but in this search coverage, have temperature anomaly (such as open-air naked light, or high temperature smog or high temperature heat source before catching fire), then the temperature fluctuation feature extracted in these temperature anomaly positions of described characteristic extracting module shows as temperature jump.The temperature fluctuation feature is the feature of temperature variation.Described characteristic extracting module passes to the alert service module with the unusual fluctuations situation of temperature, in time sends warning by the alert service module.
In the prior art, in order to hold distance, air humidity to the impact of temperature-measuring results, the common way of forest fires early warning system is on the basis of lot of experimental data, set up about the n rank matched curve of temperature-distance, temperature-humidity and corresponding error scale factor matched curve and try to achieve experimental formula by the mathematical modeling instrument, according to this to the surface temperature of object carry out about the distance, air humidity error compensation, thereby build one based on the threshold formula warning system of physics dimension (temperature).For theoretical and academic research, this can yet be regarded as is more rigorous and can a justify oneself method, but this method is difficult to satisfy the demand of practical application, because (1) in Jungln Jam, mountain top, unattended field and large space monitoring range, apart from itself being exactly a very difficult unknown factor that automatically records, so be difficult to describe with the variation of distance the variation of temperature; (2) experimental data (sample space) is difficult to reflect all sidedly that the region changes (south, the north, eastern region, west area etc.), seasonal variations (spring, summer, autumn and winter) and meteorologically changes (wind spray dust sand etc.), so be based upon matched curve on the limited experimental data basis and the limitation of experimental formula is apparent.The utility model is thought: (1) within considerable time, various for the forest fires early warning, be difficult to the demand that universality ground satisfies practical application based on the threshold formula warning system of physics dimension because of the interference that is subject to too many uncertain factor; (2) find for forest fires, temperature value itself is unimportant, importantly fluctuation relation and the fluctuation characteristic thereof between the interior temperature value of uncooled ir hygrosensor field angle (such as 320 * 240 arrays).On this understanding basis, the utility model has turned to the non-threshold formula warning system of analyzing based on to fluctuation relation and fluctuation characteristic thereof between temperature value in 320 * 240 arrays with notice from the threshold formula warning system based on physics (temperature) dimension, its basic ideas are: after having collected a frame (320 * 240 array) data, data are not relied on the temperature fluctuation feature extraction of the changed factors such as distance, atmospheric transmissivity, air humidity, wind-force.Alarm Server is nucleus equipment of the present utility model, and the sampling rate that generally is not less than 2 frames with per second gathers the result of detection of the uncooled ir hygrosensor that imports into via described network in real time, then realize above-mentioned, to the extraction of temperature fluctuation feature.
As an embodiment, described characteristic extracting module can be extracted the temperature fluctuation feature from the temperature data in the detective field of view angle of uncooled ir hygrosensor acquisition by Sobel Operator.What the temperature fluctuation feature extracting method of fluctuation characteristic extraction module adopted is that template size is 3 * 3 Suo Beier (Sobel) discrete type operator, and this method can extract the part data that temperature fluctuation has occured well from array.Sobel Operator (Sobel operator) is one of operator during image is processed, mainly as rim detection.Technically, it is a discreteness difference operator, can use this operator in order to detect and extract the profile of high-temperature area in the array in any point in 320 * 240 arrays, just can extract thus which type of sudden change has occured temperature wherein in the detective field of view angle, and whether high-temperature area the information such as is spreading.Certainly adopt Sobel Operator just to extract a kind of method of temperature fluctuation feature from temperature data, in fact in the prior art known to those skilled in the art, can may be used to characteristic extracting module from the method that temperature data extracts the temperature fluctuation feature.Therefore, these can extract the method for temperature fluctuation feature from temperature data also should be within claim protection domain of the present invention.
As the preferred embodiment of the present embodiment, data acquisition module can be the RJ45 Ethernet interface, and the sampling rate that is not less than 2 frames with per second gathers the result of detection of the uncooled ir hygrosensor that imports into via described network in real time.The alert service module can be embedded intelligent equipment or desk-top computer, comprises various communications protocols, can require and the warning client according to multiple communications mode and communication protocol, interconnects such as the third party system; Can also store, show data; All right deal with data is convenient to monitor staff or other third party's system queries real time data or historical data.Characteristic extracting module also can be embedded intelligent equipment or server or desk-top computer.
As preferred embodiment of the present utility model, shown in Fig. 1 b, described Alarm Server can also comprise characteristics analysis module, so that described characteristic extracting module is connected with described alert service module by described characteristics analysis module, the data that are characteristic extracting module are transferred to first characteristics analysis module, after characteristics analysis module is analyzed judgement to the data feature, report to the police if need, then judged result is transferred to the alert service module, gives the alarm by the alert service module.Described characteristics analysis module judges whether the current temperature fluctuation feature of surveying the moment and a upper detection moment has similarity, judge namely whether the uncooled ir hygrosensor that obtains via characteristic extracting module has obvious variation these two characteristics of surveying constantly detection, if obvious variation has occured in feature, can also be with the volume coordinate of temperature generation significant change as parameter record in addition; If two temperature fluctuation feature dissmilarities of surveying constantly, namely significant change is arranged (is the variation of feature to two temperature profiles of surveying a certain monitoring position constantly or location (this position or zone generally are the positions that occurrence temperature is suddenlyd change in characteristic extracting module) herein, but not the sudden change of temperature), because two intervals of measuring the moment are very short, therefore illustrate that fast rise has occured temperature herein, it might be open-air naked light, but also might be non-forest fires (open-air naked light) class heat radiations, such as sun lights on the rock or also may cause the fast rise of temperature on the ice face, in order to judge whether really to have occured the condition of a fire, need to continue to judge whether the dissmilarity of described temperature fluctuation feature is derived from non-forest fires (open-air naked light) class heat radiation, if not be derived from the class heat radiation, then judged result is sent to the alert service module.Signature analysis is that the variation of feature is analyzed.In this case, the raw data (such as 320 * 240 arrays) that described Alarm Server at first will stem from the uncooled ir hygrosensor imports to characteristic extracting module, judge whether to have occured the temperature anomaly fluctuation, when the temperature anomaly fluctuation has occured after, characteristics analysis module in the Alarm Server will be screened the unusual fluctuations of temperature further, to determine whether this temperature fluctuation needs to report to the police.Characteristics analysis module can be desk-top computer.If characteristic extracting module is server, characteristics analysis module can be connected with described characteristic extracting module by physical connection mode or internetwork connection mode.
The problem that described characteristics analysis module will solve is: whether (1) current time and upper a fluctuation characteristic constantly has similarity; (2) if dissimilar, also need further to differentiate this dissmilarity and whether do not belong to the open-air naked light that needs are reported to the police.A kind of method of discrimination is: have non-forest fires (open-air naked light) the class heat radiation of fixed position for kitchen smoke, the stone that has been drying heat etc., can distinguish whether its heat radiation needs to report to the police by the locus of infrared source; Non-forest fires (open-air naked light) the class heat radiation that does not have the fixed position for car engine and tire etc. can needing determine whether by the attribute (road, river, village, forest land etc.) of judging its current location affiliated area reporting to the police.(3) there is only one's own sample fluctuation feature database each visual field of each monitoring point.So both can reduce significantly the calculated amount of comparison calculation, and can make again to screen to have more specific aim.
That is to say, if two temperature fluctuation feature dissmilarities of surveying constantly, described " whether the dissmilarity of judging described temperature fluctuation feature is derived from the class heat radiation " can compare and realize by temperature fluctuation feature and the sample characteristics in the sample fluctuation feature database (can be to be preset in the sample fluctuation feature database or real-time update) that will extract.Described sample fluctuation feature database has been stored the fluctuation characteristic data of different geographic entitys, the temperature fluctuation feature that has separately self such as river, road, kitchen smoke etc., can also store the sample of the temperature fluctuation feature of a certain concrete coordinates regional, such as first will be under normal circumstances the temperature fluctuation feature in certain section forest zone as sample storage, the temperature fluctuation feature in this section forest zone that again real-time detection is obtained by comparison, if similar, illustrate and be not the condition of a fire, if dissimilar, then very likely would be the condition of a fire.In this case, the workflow of Alarm Server as shown in Figure 2.Described sample fluctuation feature database can be the database that is connected with described characteristics analysis module.
For other the non-forest fires that can't determine in advance (open-air naked light) class heat radiations, the ability of self-teaching that can have by sample fluctuation feature database (in the utility model also referred to as Sample Storehouse) little by little accumulates samples of undulations.The ability of self-teaching of Sample Storehouse stems from true newspaper, reports by mistake and fails to report the temperature fluctuation feature that system automatically records in the event.Specifically, for miss/failing to report event (generally speaking, the utility model does not allow to occur failing to report event), Sample Storehouse can automatically record a stack features value (position angle of detector, volume coordinate, environment temperature, wind-force, wind direction, generation miss/fail to report event date/time and block attribute, cause the high-temperature area edge feature miss/failed to report and/or temperature fluctuation feature etc.).This ability of self-teaching of Sample Storehouse is so that whole system can be improved constantly and constantly improve examination ability for various particular thermal radiation sources.
If two temperature fluctuation feature dissmilarities of surveying constantly, characteristics analysis module described " continuing to judge whether the dissmilarity of temperature fluctuation feature is derived from the class heat radiation " (being the another kind of determination methods of characteristics analysis module), by with the temperature fluctuation feature of extracting recently realization relative to the temperature fluctuation feature of previous moment same space coordinate, if the significant change of the temperature fluctuation feature displays temperature of this moment constantly enlarges in certain spatial dimension, illustrate that very likely the intensity of a fire is spreading, should report to the police as early as possible this moment.This determination methods not only can be carried out the forest fires early warning, and can be after the condition of a fire occurs, the situation that spreads of the monitoring intensity of a fire.Above-mentioned two kinds of determination methods can certainly be combined, screen more accurately infrared source whether the condition of a fire has occured.
As preferred embodiment, uncooled ir hygrosensor in the forest fires early warning system of the present utility model can also have some annexes, such as comprising digital tripod head, each described uncooled ir hygrosensor is connected with digital tripod head, digital tripod head plays a supportive role to the uncooled ir hygrosensor, but also can carry out Remote by described network to it by Alarm Server or other terminal, with detection angle, direction, the distance of adjusting uncooled ir hygrosensor, and the setting of cruising etc.Usually, digital tripod head can be set to entirely cruise within sweep of the eye at spheroid according to certain hour.This has just solved and has been fixedly connected with in the situation, and the uncooled ir hygrosensor can only be surveyed, checks and transmit the data message of certain deflection and can not look about from all directions shortcoming centered by it.Can also comprise, in order to resist severe weather conditions, and the uncooled ir hygrosensor is packaged in wherein protective cover.Protective cover can be installed on the digital tripod head, to realize the scanning of cruising to the forest zone.
In actual applications, preferably, the uncooled ir hygrosensor can link to each other with network communication and transmission equipment by the RJ-45 network interface, to realize the remote network control to the uncooled ir hygrosensor by network communication and transmission equipment.
For the ease of expressing, uncooled ir hygrosensor, detector protective cover, digital tripod head and network communication and transmission equipment are generically and collectively referred to as the front-end dynamic forest fire prior-warning device.
As a preferred embodiment of the present utility model, described Alarm Server can be connected with the warning client, and then described Alarm Server can be connected by Wide Area Network with the warning client, shown in Fig. 1 b.Described warning client can be forest fire protection alarm and command systems at different levels and/or third party system, such as meteorological system, forest fires satellite monitoring system, forest ranger system, Forest Police system, forest People's Armed Police system, forest fires put out a fire to save life and property emergency commading system, Geographic Information System etc., described Alarm Server can provide alert service to described warning client by following three kinds of modes: 1) SDK mode: provide API (application development interface) to the warning client; 2) message subscribing manner: the warning client can get access to warning message by message subscribing manner; 3) message push mode: can be on one's own initiative to warning client push warning message.
Preferably, for on Alarm Server and/or the warning client show the infrared imaging data, can also comprise encoder, described scrambler is connected with the uncooled ir hygrosensor, the infrared imaging data are encoded, described demoder is connected with Alarm Server and/or warning client, to the infrared imaging decoding data of having encoded.In this case, described scrambler also belongs to the front-end dynamic forest fire prior-warning device.In order to realize the early warning to forest fires in the large space scope, the uncooled ir hygrosensor can be placed on the commanding elevation of mountain forest usually, away from (county, districts and cities, province etc.) at different levels Forest Fire Monitoring center.So, the uncooled ir hygrosensor can link to each other with scrambler (a kind of embedded image compression and network communication and transmission equipment), and infrared imaging data etc. are sent to Alarm Server and/or warning client (such as the third party system etc.) via Wide Area Network (normally Wireless microwave network or telstar).Certainly, scrambler is as the instrument of network-in-dialing, and the temperature data that also the uncooled ir hygrosensor can be collected is sent to Alarm Server and/or warning client by Wide Area Network.
Preferably, described digital tripod head can pass through network interface (such as RS-485) and link to each other with scrambler, to realize the remote network control to digital tripod head by scrambler, shown in Fig. 1 b.
Preferably, described uncooled ir hygrosensor links to each other with scrambler by bnc interface, to realize the compressed encoding to thermal-induced imagery (being above-mentioned infrared imaging data), shown in Fig. 1 b.Scrambler is transferred to Forest Fire Monitorings at different levels center by the thermal-induced imagery of its network transmission module after with compressed encoding.
Preferably, described scrambler links to each other with network communication and transmission equipment by the RJ-45 interface, shown in Fig. 1 b, to realize the network interconnection at front-end dynamic forest fire prior-warning device and Forest Fire Monitorings at different levels center.
Alphanumeric on the line among Fig. 1 a and Fig. 1 b is the example of the line in the embodiment of the present utility model; just to explanation of the present utility model; those skilled in the art understand the line can use other form and it is replaced and change, so the concrete form of these lines should not thought the restriction to the utility model protection domain.
In the utility model, the relation between described Alarm Server and uncooled ir hygrosensor is the MS master-slave relation.The running parameter of uncooled ir hygrosensor is pre-determined and is adjusted by Alarm Server; Alarm Server is according to parameters such as the capacity of the network planning and Internet Transmission channel, quality, determines the network operation parameters such as the network address, network transmission speed, image frame per second of uncooled ir hygrosensor.
A kind of forest fires early warning system based on the uncooled ir hygrosensor of the present utility model, comprise the bottom Alarm Server and with at least one deck upper strata Alarm Server of described bottom Alarm Server by network connection, described bottom Alarm Server is the Alarm Server of above-mentioned forest fires early warning system.Described bottom Alarm Server can interconnecting by network interface (such as the RJ-45 network interface) realization and front-end dynamic forest fire prior-warning device.Described upper strata Alarm Server can carry out classification according to management area and/or supervisory level and dispose (province, city, county, township etc.), is arranged to the upper strata Alarm Server of multilayer.All front-end dynamic forest fire prior-warning devices in all bottom Alarm Server compasss of competency all can automatically be registered to the upper strata Alarm Server, thereby realize that Alarm Server is to the data acquisition of each the front-end dynamic forest fire prior-warning device in its compass of competency, can avoid because the warning message that a variety of causes (such as carelessness or unmanned or the erroneous judgement etc. of certain one-level forest fires early warning unit) causes is out in the cold or delay, thereby can to the full extent and in the fastest time warning message be sent to competent authorities, reduce to the full extent the loss of national wealth.This shows, a plurality of Alarm Servers have good classification and dispose ability, can be cascaded into large-scale forest fires early warning system, can certainly dispose separately for certain specific demand, only adopt the form of or one deck bottom Alarm Server.
Alarm Server in the forest fires early warning system of the present utility model has solved the difficult problem that the threshold formula warning system based on physics (temperature) dimension is faced with in actual applications effectively by characteristic extracting module and characteristics analysis module, thereby realized in the large space scope, under non-definite distance and the meteorological condition, to round-the-clock monitoring and the early warning of forest fires.(1) the Alarm Server sampling rate that can per second be not less than 2 frames by data acquisition module gathers the result of detection of the uncooled ir hygrosensor that imports into via described network in real time; (2) by extract, the analysis temperature fluctuation characteristic, solved the uncooled ir hygrosensor in the large space scope, because being subject to the impact of the factors such as detection range, air humidity, and be difficult to set the problem of alarming threshold value; The unusual fluctuations of the temperature that (3) non-forest fires (open-air naked light) the class heat radiations such as car engine and tire, kitchen smoke is caused have good examination ability.
As an embodiment, the step that the utility model is realized the forest fires early warning as shown in Figure 2: the raw data that (1) collects the uncooled ir hygrosensor via described network delivery to Alarm Server.(2) by the characteristic extracting module of Alarm Server the raw data that collects is processed processing, form the fluctuation characteristic data.(3) the fluctuation characteristic analysis module is at first differentiated the unusual fluctuations whether temperature has occured.(4) if the temperature anomaly fluctuation has occured, whether the unusual fluctuations that characteristics analysis module is further differentiated temperature stem from non-forest fires (open-air naked light) class heat radiation, thereby determine whether need to report to the police.(5) the alert service module can provide alert service to the warning client in several ways.Described forest fires early warning system can be in unattended situation, and Real-Time Monitoring and early detection to forest fires (open-air naked light) were realized in ground in the large space scope in 7 * 24 hours.
The utility model has successfully solved the impact on uncooled ir hygrosensor temperature-measuring results such as the distance factor that can't determine in advance, air humidity factor, has realized one based on the forest fires early warning system uncooled ir hygrosensor, that practicality is good in the large space scope.Experimental result shows, the utility model is very responsive to open-air naked light, and to non-forest fires (open-air naked light) class heat radiations such as car engine and tire, kitchen smoke, has good examination ability.
Claims (7)
1. forest fires early warning system based on the uncooled ir hygrosensor, it is characterized in that, comprise at least one uncooled ir hygrosensor, network, Alarm Server, described uncooled ir hygrosensor and described Alarm Server pass through respectively network communication and transmission equipment and described network connection separately.
2. forest fires early warning system according to claim 1, it is characterized in that, described Alarm Server comprises data acquisition module, characteristic extracting module and the alert service module that links to each other successively, and described data acquisition module is by the temperature data in the detective field of view angle of described network collection uncooled ir hygrosensor and be transferred to characteristic extracting module; Described characteristic extracting module passes to the alert service module with the unusual fluctuations situation of temperature, in time sends warning by the alert service module.
3. forest fires early warning system according to claim 2 is characterized in that, described Alarm Server also comprises characteristics analysis module, and described characteristic extracting module is connected with described alert service module by described characteristics analysis module.
4. according to claim 1 to one of 3 described forest fires early warning systems, it is characterized in that, also comprise digital tripod head, each described uncooled ir hygrosensor is connected with digital tripod head.
5. according to claim 1 to one of 3 described forest fires early warning systems, it is characterized in that, also comprise the warning client, described Alarm Server is connected by Wide Area Network with the warning client, and providing alert service to the warning client, described warning client is forest fire protection alarm and command systems at different levels and/or third party system.
6. forest fires early warning system according to claim 5 is characterized in that, also comprises encoder, and described scrambler is connected with the uncooled ir hygrosensor, and described demoder is connected with Alarm Server and/or warning client.
7. forest fires early warning system based on the uncooled ir hygrosensor, it is characterized in that, comprise the bottom Alarm Server and with at least one deck upper strata Alarm Server of described bottom Alarm Server by network connection, described bottom Alarm Server is the Alarm Server of the described forest fires early warning system of one of claim 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201220069019 CN202650186U (en) | 2012-02-28 | 2012-02-28 | Forest fire early warning system based on uncooled infrared temperature detector |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201220069019 CN202650186U (en) | 2012-02-28 | 2012-02-28 | Forest fire early warning system based on uncooled infrared temperature detector |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN202650186U true CN202650186U (en) | 2013-01-02 |
Family
ID=47419256
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201220069019 Expired - Fee Related CN202650186U (en) | 2012-02-28 | 2012-02-28 | Forest fire early warning system based on uncooled infrared temperature detector |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN202650186U (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103793901A (en) * | 2014-01-26 | 2014-05-14 | 上海热像机电科技有限公司 | Infrared thermal image system supporting real-time panoramic stitching of total-radiation infrared thermal image video streaming |
-
2012
- 2012-02-28 CN CN 201220069019 patent/CN202650186U/en not_active Expired - Fee Related
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103793901A (en) * | 2014-01-26 | 2014-05-14 | 上海热像机电科技有限公司 | Infrared thermal image system supporting real-time panoramic stitching of total-radiation infrared thermal image video streaming |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102881109B (en) | Forest fire early-warning system based on non-refrigerant infrared temperature detector | |
| Oliva et al. | Assessment of VIIRS 375 m active fire detection product for direct burned area mapping | |
| CN205582179U (en) | Forest fire remote monitoring system based on satellite remote sensing image | |
| US7991187B2 (en) | Intelligent image smoke/flame sensor and detection system | |
| Molina-Pico et al. | Forest monitoring and wildland early fire detection by a hierarchical wireless sensor network | |
| CN110491066A (en) | Forest fire protection monitoring and warning system based on infrared thermal imaging | |
| CN103455708A (en) | Power transmission line disaster monitoring and risk assessment platform based on satellite and weather information | |
| CN114724337B (en) | Remote intelligent monitoring and early warning system and method based on photovoltaic cell power supply | |
| CN109448292A (en) | A kind of power grid mountain fire monitoring and pre-alarming method | |
| CN115348247A (en) | Forest fire detection early warning and decision-making system based on sky-ground integration technology | |
| CN104581076A (en) | Mountain fire monitoring and recognizing method and device based on 360-degree panoramic infrared fisheye camera | |
| CN106910311B (en) | A kind of real-time wide area synchronous satellite monitoring method of power network mountain fire and monitoring system | |
| CN102646311A (en) | Intelligent smoke and fire detecting system using real-time dynamic cruising images | |
| CN108806158A (en) | A kind of forest fire detection rescue system based on cloud computing | |
| CN104007728A (en) | Digital city sensing device, system and method | |
| CN112750280B (en) | Combined type geological disaster professional monitoring and early warning method and device | |
| CN104613892A (en) | Video detection technology and laser ranging technology integrated compound snow depth monitoring system | |
| CN205354263U (en) | Forest fire detection system based on infrared thermal imagery | |
| CN103514702A (en) | Forest fire prevention early warning system | |
| CN201091014Y (en) | Fire detecting device | |
| WO2015103719A1 (en) | Fire monitoring system | |
| CN106971510A (en) | A kind of double mode mountain torrents pre-alarming control system and method | |
| CN105448017A (en) | Community rail alarming system | |
| CN114374716A (en) | Geological disaster remote monitoring system and monitoring method thereof | |
| Kolarić et al. | Integrated system for forest fire early detection and management |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 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: 20130102 Termination date: 20150228 |
|
| EXPY | Termination of patent right or utility model |