CN203299842U - Thermal infrared image face recognition system adapted in underground coal mine - Google Patents
Thermal infrared image face recognition system adapted in underground coal mine Download PDFInfo
- Publication number
- CN203299842U CN203299842U CN2013202745518U CN201320274551U CN203299842U CN 203299842 U CN203299842 U CN 203299842U CN 2013202745518 U CN2013202745518 U CN 2013202745518U CN 201320274551 U CN201320274551 U CN 201320274551U CN 203299842 U CN203299842 U CN 203299842U
- Authority
- CN
- China
- Prior art keywords
- thermal infrared
- face
- module
- aboveground
- infrared images
- 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
- 239000003245 coal Substances 0.000 title claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 42
- 239000013598 vector Substances 0.000 claims abstract description 39
- 238000004891 communication Methods 0.000 claims abstract description 33
- 238000005065 mining Methods 0.000 claims abstract description 18
- 238000003860 storage Methods 0.000 claims abstract description 14
- 230000005540 biological transmission Effects 0.000 claims description 13
- 238000005070 sampling Methods 0.000 claims description 11
- 238000012546 transfer Methods 0.000 claims description 8
- 230000002123 temporal effect Effects 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 31
- 238000009826 distribution Methods 0.000 abstract description 11
- 238000001514 detection method Methods 0.000 abstract description 9
- 238000005286 illumination Methods 0.000 abstract description 6
- 230000007613 environmental effect Effects 0.000 abstract description 5
- 239000010883 coal ash Substances 0.000 abstract description 2
- 210000004204 blood vessel Anatomy 0.000 description 25
- 239000011521 glass Substances 0.000 description 14
- 230000001815 facial effect Effects 0.000 description 13
- 230000000877 morphologic effect Effects 0.000 description 13
- 238000001914 filtration Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 8
- 239000000284 extract Substances 0.000 description 8
- 238000009792 diffusion process Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000000605 extraction Methods 0.000 description 4
- 238000001757 thermogravimetry curve Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 239000000306 component Substances 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 239000008358 core component Substances 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 244000287680 Garcinia dulcis Species 0.000 description 1
- 238000007630 basic procedure Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003818 cinder Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000002834 transmittance Methods 0.000 description 1
Images
Landscapes
- Image Processing (AREA)
Abstract
The utility model discloses a morphological-method-based thermal infrared image face recognition system adapted in an underground coal mine, which comprises a processing unit, an infrared thermal imager, a mining temperature sensor, a storage unit, a communication line, an aboveground server and an aboveground attendance database. The processing unit is used for carrying out the recognition operation. The infrared thermal imager is used for acquiring the thermal infrared images of human faces. The temperature sensor is used for acquiring the environmental temperature to carry out the temperature compensation on the above images. The storage unit is used for storing the feature vectors of the cross-point distribution of available miners. The communication line is used for transmitting a recognition result to the aboveground server. According to the recognition result, the aboveground server sends out a corresponding warning or an attendance judging decision. The aboveground attendance database is used for recording the information of miners going in or out of an underground coal mine. According to the utility model, by means of the system, human faces are recognized based on thermal infrared images. The influences of poor illumination condition in an underground coal mine and coal ashes on the faces of miners can be effectively overcome. Therefore, the thermal infrared image face recognition system is good in reliability and high in recognition rate, and provides reliable information for the detection on the identities of personnel going in or out of a coal mine in a mining area.
Description
Technical field
The utility model relates to a kind of face identification system of thermal infrared images based on morphological image, belongs to the image pattern recognition field.
Background technology
Coal is one of main energy of China, is the important leverage of China's economic development, China approximately 70% energy resource consumption from coal., because miner's violation, operation against rules happen occasionally, can affect the safety in production under coal mine, thereby in the downhole production process, the personal management of going into the well be had strict requirement and restriction.
Going out, enter the detection of well personnel identity is to guarantee one of important technical of mine safety production.Present existing detection method comprises the detection technique based on RFID and fingerprint, these technology can play effect to a certain extent, but also there are some problems, detect and need the miner to wear radio-frequency identification card as the personnel based on the RFID radio-frequency technique, but radio-frequency identification card is not the indivisible recognition feature of miner: the miner can carry a plurality of other miners' radio-frequency card, has the phenomenon for brush for the people; A plurality of miners also can share a radio-frequency card, exist non-miner to flee to enter the phenomenon in mining area; Intercourse radio-frequency identification card between the miner, have the phenomenon of changing privately work.Utilize people's finger print information identification personnel object based on the detection technique of fingerprint, although finger print information is the biological characteristic of human body self, possess and human body between good indivisible characteristic, but because the miner is engaged in the manual labor process, finger usually speckles with cinder or finger is often injured in the course of the work, and this all makes the fingerprint recognition rate low.
Aboveground recognition technology based on biological characteristic also comprises face recognition technology at present, and face recognition technology is divided into the visible light recognition of face, near infrared recognition of face and thermal infrared recognition of face.Aboveground visible light face recognition technology is widely used aspect recognition detection, comparatively ripe method has face identification method based on eigenface, based on the face identification method of linear discriminant analysis etc., but various visible light face recognition algorithms have comparatively strict requirement for the illumination condition in environment-identification, in illumination variation and dim environment, the correct recognition rata of system will reduce greatly.Personnel to be detected face is when covering ornament in addition, and recognition effect also can be greatly affected.Because the down-hole illumination condition is poor, and the miner works rear face with a large amount of coals stains,, although the visible light face recognition technology is comparatively ripe, utilizes the visible light face recognition technology still can not reach good recognition effect in subsurface environment.Near infrared recognition of face and thermal infrared recognition of face are in the starting stage of research at present, and the near infrared recognition of face is less demanding to illumination, but need people's face especially near near-infrared image forming apparatus.Obtaining of thermal infrared images is comparatively simple, what the image that obtains reflected is the heat distribution situation on people's face surface, even can not have influence on image quality under illuminance utmost point low condition yet, and the ornament that people's face surface exists can not cause too much influence to imaging yet, and the thermal infrared facial image is particularly suitable under coal mine to realize the identification to people's face.Present aboveground thermal infrared face identification method comprises is transplanted to recognition methods in the thermal infrared recognition of face (as principal component analytical method with the visible light face identification method, the linear discriminant analysis method), based on morphologic recognition methods, these methods have all obtained certain recognition effect.
Because the colliery down-hole ambient temperature is different from surface temperature, and the workman is through high-intensity operation, tend to cause the variation of people's face surface temperature, Coal Miners also can be worn mine lamp in addition, mine lamp will be higher than people's face temperature in the rear surface temperature that works long hours, have influence on the recognition effect to people's face, therefore in the process of carrying out the thermal infrared recognition of face, need to overcome the impact of these factors on recognition effect.Because glass is poor to infrared transmittance, identify for people's face of wearing spectacles is also one of difficult point of thermal infrared recognition of face always.The present invention utilizes the identification of thermal infrared images realization to the coal mine down-hole face, has proposed detection and disposal route to wearing mine lamp and glasses, utilizes temperature compensation to reduce the impact of the Factors on Human face discriminations such as variation of ambient temperature.
Summary of the invention
In order to overcome the problem of existing RFID tag card method, fingerprint identification method, existence when the visible light face identification method is applied to down-hole uniqueness detection, the utility model discloses a kind of face identification system of thermal infrared images based on morphological method, system can reach higher discrimination in the dim environment in down-hole,, to variation of ambient temperature strong adaptability under coal mine, for going out, enter the detection of well personnel uniqueness, bank provides guarantee.
Thermal infrared images based on morphological method face identification system described in the utility model is by processing unit, thermal infrared imager, and mining temperature sensor, storage unit, communication line, aboveground server and aboveground attendance data storehouse form.
Described processing unit comprises control module, recognition of face and processing module, proper vector access module and communication module; Wherein control module comprises again temperature acquisition control module and thermal infrared images controlling of sampling module; Processing unit is the core component of whole face identification system, and it mainly contains the function of three aspects: (1) is to system peripherals equipment: the control of thermal infrared imager and mining temperature sensor; (2) to gathering processing and the identification of facial image; (3) to aboveground server transmission recognition result;
Described temperature acquisition control module is used for controlling described mining temperature sensor,, with certain sampling period collection environment temperature, with people's face thermal infrared images that described thermal infrared imager is collected, carries out temperature compensation; Described thermal infrared images controlling of sampling module is used for controlling described thermal infrared imager collection people face thermal infrared images; Described recognition of face and processing module are used for obtaining people's face identity; Described proper vector access module is used for the crossover points distributed proper vector that access miner face characteristic storehouse establishment stage obtains; Described communication module is used for that described recognition of face is obtained result with the processing module processing and transfers to described aboveground server.
Described thermal infrared imager is a kind of equipment that utilizes infrared eye and optical imagery object lens to accept the infrared energy of measured target, and obtain the Infrared Thermogram of reflect heat energy distribution, Infrared Thermogram is corresponding with the heat distribution field of body surface, is used for gathering people's face thermal infrared images under the control of described processing unit; Described mining temperature sensor is selected the explosion-proof type temperature sensor, is used for gathering current environmental temperature information under the control of described processing unit, with people's face thermal infrared images that described thermal infrared imager is collected, carries out temperature compensation; Described storage unit, be used for all miners that storage miner's face characteristic storehouse establishment stage obtains and wearing or the crossover points distributed proper vector under wearing spectacles and mine lamp condition not; Described communication line, under the control of processing unit communication module, utilize asynchronous serial I/O mode to carry out the transmission of data, one of each character of the transmission of data is connect a transmission, data each not coordination can timesharing use same transmission channel to transmit, line efficiency is high, and minimum only the need can communicate with a pair of circuit, and communication line is passed to aboveground server with processing unit to the recognition result information of inputting people's face thermal infrared images to be identified; Described aboveground server comprises aboveground server end communication module, work attendance module and alarm module; Aboveground server end communication module is responsible for communicating by letter with the communication module of down-hole treatment unit, receives the recognition result to people's face to be identified; The work attendance module is used for recognition result is write described aboveground attendance data storehouse; Alarm module is used for warning aboveground managerial personnel; Aboveground server is used for receiving and distributing the recognition result that the down-hole treatment unit comes by the communication line transmission; Described aboveground attendance data storehouse, for record discrepancy well miner's identity, temporal information.
The described face identification system of thermal infrared images based on morphological method workflow is as follows:
Miner's face characteristic storehouse establishment stage, at first system extracts the crossover points distributed proper vector after all sequences in people's face thermal infrared images that all miners wore or do not wear mine lamp or glasses at ambient temperature, and these crossover points distributed proper vectors are deposited in miner's face characteristic storehouse by the proper vector memory module;
In people's face ONLINE RECOGNITION stage, after the normal startup of system, operational scheme is as follows:
(1) by control module initialization mining temperature sensor and thermal infrared imager;
(2) thermal infrared images controlling of sampling module controls thermal infrared imager gathers people's face thermal infrared images, and detects in the image that collects whether comprise facial image, if inclusion test, to human face region, arrives recognition of face and processing module with image transfer;
(3) the proper vector memory module reads crossover points distributed proper vector corresponding to each miner's facial image in miner's face characteristic storehouse;
(4) recognition of face and processing module recipient face thermal infrared images, and people's face thermal infrared images is processed and obtained the crossover points distributed proper vector that comprises in people's face thermal infrared images to be identified, according to obtaining recognition result to matching degree between miner's face characteristic storehouse chats workman face crossover points distributed proper vector and miner people's face crossover points distributed proper vector to be identified;
(5) communication module transfers to aboveground server with the recognition result of recognition of face and processing module, and aboveground server is made respective handling according to recognition result,, if detect the unauthorized miner of going into the well of people's face to be identified, to aboveground control center, gives a warning; If detect go into the well miner coupling of people's face to be identified and certain mandate, the miner's that will identify personal information, current time information are recorded to aboveground attendance data storehouse;
The beneficial effects of the utility model are, utilize thermal infrared images identification people face, can effectively overcome the down-hole illumination condition poor, and there is the impact of coal ash in underground work personnel face.The recognition system reliability is strong, and discrimination is high, and strong adaptability detects for mining area goes out, enters well personnel identity uniqueness the reliable information that provides.
Description of drawings
The present invention is described in further detail below in conjunction with the drawings and specific embodiments.
Fig. 1 is based on the thermal infrared images face identification system structural drawing of morphological method;
Fig. 2 is each module workflow diagram of system processing unit;
Fig. 3 is based on thermal infrared images recognition of face process flow diagram under the coal mine of morphological image;
In figure, 1, processing unit; 2, thermal infrared imager; 3, mining temperature sensor; 4, storage unit; 5, communication line; 6, aboveground server; 7, aboveground attendance data storehouse;
Embodiment
At first the structure of the thermal infrared images face identification system based on morphological method is described.With reference to Fig. 1, described thermal infrared images face identification system is by processing unit 1, thermal infrared imager 2, and mining temperature sensor 3, storage unit 4, communication line 5, aboveground server 6 and database 7 form.
Described processing unit 1 comprises control module, recognition of face and processing module, proper vector access module and communication module; Wherein control module comprises again temperature acquisition control module and thermal infrared images controlling of sampling module; Described processing unit 1 is the core component of whole face identification system, and it mainly contains the function of three aspects: (1) is to system peripherals equipment: the control of thermal infrared imager and mining temperature sensor; (2) to gathering processing and the identification of facial image; (3) to aboveground server transmission recognition result;
Described temperature acquisition control module is used for controlling described mining temperature sensor,, with certain sampling period collection environment temperature, with people's face thermal infrared images that described thermal infrared imager is collected, carries out temperature compensation; Described thermal infrared images controlling of sampling module is used for controlling described thermal infrared imager collection people face thermal infrared images; Described recognition of face and processing module are used for people's face thermal infrared images utilization that thermal infrared imager collects is processed and identified based on the thermal infrared images face identification method of morphological method; The crossover points distributed proper vector that described proper vector access module is used for miner's face characteristic storehouse establishment stage is obtained stores described storage unit into, perhaps from described storage unit, extracts the crossover points distributed proper vector and is used for the identification in people's face ONLINE RECOGNITION stage; Described communication module is used for that described recognition of face is obtained result with the processing module processing and transfers to described aboveground server.
Described thermal infrared imager 2 is a kind of equipment that utilizes infrared eye and optical imagery object lens to accept the infrared energy of measured target, and obtain the Infrared Thermogram of reflect heat energy distribution, Infrared Thermogram is corresponding with the heat distribution field of body surface, is used for gathering people's face thermal infrared images under the control of described processing unit; Described mining temperature sensor 3 is selected the explosion-proof type temperature sensor, is used for gathering current environmental temperature information under the control of described processing unit, with people's face thermal infrared images that described thermal infrared imager is collected, carries out temperature compensation; Described storage unit 4, be used for all miners that storage miner's face characteristic storehouse establishment stage obtains and wearing or the crossover points distributed proper vector under wearing spectacles and mine lamp condition not; Described communication line 5, under the control of processing unit communication module, utilize asynchronous serial I/O mode to carry out the transmission of data, one of each character of the transmission of data is connect a transmission, data each not coordination can timesharing use same transmission channel to transmit, line efficiency is high, and minimum only the need can communicate with a pair of circuit, and communication line is passed to described aboveground server 6 with processing unit to the recognition result information of inputting people's face thermal infrared images to be identified; Described aboveground server 6 comprises aboveground server end communication module, work attendance module and alarm module; Aboveground server end communication module is responsible for communicating by letter with the communication module of down-hole treatment unit, receives the recognition result to people's face to be identified; , according to the difference of down-hole face recognition result, when identifying miner's identity, utilize the work attendance module to carry out the work work attendance to the miner; While going out miner's identity, utilize alarm module to notify aboveground managerial personnel unidentified; Aboveground server is responsible for the recognition result that comes by the communication line transmission according to the down-hole treatment unit, make corresponding processing, if find that unauthorized miner enters well, gives a warning, authorize the miner to go out well or enter well if recognition result shows, record out, enter well information and be used for miner's work attendance of working; Described aboveground attendance data storehouse 7, be used under the control of described aboveground server, automatically records the well miner's that comes in and goes out identity, temporal information according to face recognition result.
Fig. 2 is based on the workflow of the thermal infrared images face identification system of morphological method:
Miner's face characteristic storehouse establishment stage, at first system extracts the crossover points distributed proper vector after all sequences in people's face thermal infrared images that all miners wore or do not wear mine lamp or glasses at ambient temperature, and these crossover points distributed proper vectors are deposited in miner's face characteristic storehouse by the proper vector memory module;
In people's face ONLINE RECOGNITION stage, after the normal startup of system, operational scheme is as follows:
(1) by control module initialization mining temperature sensor and thermal infrared imager;
(2) thermal infrared images controlling of sampling module controls thermal infrared imager gathers people's face thermal infrared images, and detects in the image that collects whether comprise facial image, if inclusion test, to human face region, arrives recognition of face and processing module with image transfer;
(3) the proper vector memory module reads crossover points distributed proper vector corresponding to each miner's facial image in miner's face characteristic storehouse;
(4) recognition of face and processing module recipient face thermal infrared images, and people's face thermal infrared images is processed and obtained the crossover points distributed proper vector that comprises in people's face thermal infrared images to be identified, according to obtaining recognition result to matching degree between miner's face characteristic storehouse chats workman face crossover points distributed proper vector and miner people's face crossover points distributed proper vector to be identified;
(5) communication module transfers to aboveground server with the recognition result of recognition of face and processing module, and aboveground server is made respective handling according to recognition result,, if detect the unauthorized miner of going into the well of people's face to be identified, to aboveground control center, gives a warning; If detect go into the well miner coupling of people's face to be identified and certain mandate, the miner's that will identify personal information, current time information are recorded to aboveground attendance data storehouse;
With reference to Fig. 3, the basic procedure of thermal infrared images face identification method under the coal mine based on morphological image is described, concrete steps are as follows, the establishment stage in miner's face characteristic storehouse:
(1) at room temperature condition T
houUnder, gather respectively the people face thermal infrared direct picture sample (every miner four width) of miner when wearing and not wearing mine lamp and glasses that has the authority of going into the well, and the thermal infrared images sample that collects is carried out normalized: intercept out the facial image that does not contain background, and the gray-scale value of image is transformed to interval [0,255];
(2) image after normalization is made anisotropic diffusion filtering and process, to reach the purpose that strengthens the vessel boundary feature in thermal infrared images and remove noise;
What people's face thermal infrared images reflected is the heat distribution situation on people's face surface, and the heat distribution of people's face surface is the result of heat diffusion in blood vessel.Heat transfer process produces level and smooth thermograde around making blood vessel, and what obtain in thermal infrared images is the variation of gray scale.Anisotropic diffusion filtering can be strengthened the vessel boundary that is produced by diffusion effectively in denoising, after filtering is processed, can more effectively extract blood vessel network.
The formula of the anisotropic diffusion filtering of 4 directions under discrete conditions is:
Wherein, I
tThe gray-scale value that (x, y) expression is located at point (x, y) through the image after the t time iterative filtering;
c
N, t(x, y), c
S, t(x, y), c
E, t(x, y), c
W, t(x, y) is corresponding is the spread function value that after the t time iterative filtering, the picture point (x, y) on the four direction of upper and lower, left and right is located, and the spread function expression formula is
Wherein d can get N, S, E, one of W}, represent the spread function of certain direction; T is iterations;
The maximal value that represents the gradient of image on the d direction of the t time iteration;
(3) utilize the morphological image method to extract the blood vessel network in image after filtering, obtain the blood vessel network image;
(4) extract the skeleton of single pixel wide of blood vessel network image medium vessels network;
(5) the intersecting blood vessels point in extraction blood vessel network skeleton, as the unique point that recognition of face utilizes, obtains the position (x of all intersecting blood vessels points
i, y
i) and calculate the Vessel orientation functional value Ψ of each intersecting blood vessels point position
i=ψ (x
i, y
i), i=1 wherein, 2 ..., M, M presentation video medium vessels point of crossing sum;
Utilize the intersecting blood vessels point (x of all extractions
i, y
i, Ψ
i), i=1,2 ..., M, calculate the crossover points distributed proper vector that reflects each intersecting blood vessels point distribution situation in image: V
i={ (d
1, φ
1), (d
2, φ
2) ..., (d
N, φ
N), Ψ
i, i=1,2 ..., M, wherein, d
1, d
2..., d
NExpression point of crossing (x
i, y
i, Ψ
i) and the distance value between nearest N point of crossing with it, φ
1, φ
2..., φ
NExpression point of crossing (x
i, y
i, ψ
i) and the line between nearest N point of crossing and the angle between the image level direction with it.In the image normalization process, can not guarantee that all intersecting blood vessels point absolute positions in image remain unchanged, substantially constant but the relative position between point of crossing but can keep, so V
iI point of crossing (x can better be described
i, y
i, Ψ
i) position feature in facial image.The point of crossing number that generally can extract from facial image is in the 20-30 scope, and the N that therefore can get in the crossover points distributed proper vector is 10.
When gathering the miner and wear image under mine lamp (or glasses) state, at first detect the position of mine lamp (or glasses), obtain position and cover plate, then utilize position to cover the plate overlay image, to extract the intersecting blood vessels point except mine lamp (or glasses) blocks position, and then obtain crossover points distributed proper vector V
i, i=1,2 ..., M;
(6) to the top n component (d of all crossover points distributed proper vectors
1, φ
1), (d
2, φ
2) ..., (d
N, φ
N) respectively according to φ order sequence from small to large, the vectorial SV after being sorted
i={ (d
(1), φ
(1)), (d
(2), φ
(2)) ..., (d
(N), φ
(N)), Ψ
i, i=1,2 ..., M, wherein φ
(1)＜φ
(2)＜...＜φ
(N)Finally with SV
i, i=1,2 ..., M stores miner's face characteristic storehouse into, is used for people's face ONLINE RECOGNITION stage;
People's face ONLINE RECOGNITION stage:
(1) gather miner's to be identified thermal infrared images, utilize temperature sensor to obtain current environmental temperature T
erv
(2) the miner's thermal infrared facial image that collects is done normalized: intercept out the facial image that does not contain background, and the gray-scale value of image is transformed to interval [0,255];
(3) whether the people's face in detected image wears mine lamp or glasses, and obtains the position illiteracy plate of mine lamp or glasses;
(4) according to current environmental temperature T
ervRoom temperature T while with miner's face characteristic storehouse establishment stage, gathering image
houBetween the facial image of difference after to normalization carry out temperature compensation;
(5) image after temperature compensation is made anisotropic diffusion filtering and process, strengthen the vessel boundary feature in thermal infrared images and remove noise;
(6) utilize the morphological image method to extract the blood vessel network in image after filtering, obtain the blood vessel network image;
(7) extract the skeleton of single pixel wide of blood vessel network image medium vessels network;
(8) the intersecting blood vessels point in extraction blood vessel network skeleton, as the unique point that recognition of face utilizes, obtains the position (x of all intersecting blood vessels points
i, y
i) and calculate the Vessel orientation functional value Ψ of each intersecting blood vessels point position
i=ψ (x
i, y
i), i=1 wherein, 2 ..., M, M represent miner's to be identified thermal infrared images medium vessels point of crossing sum; Utilize the intersecting blood vessels point (x of all extractions
i, y
i, Ψ
i), i=1,2 ..., M, calculate the crossover points distributed proper vector V that reflects each intersecting blood vessels point distribution situation in image
i, i=1,2 ..., M; If current collection is the image of miner while wearing mine lamp (or glasses), utilize position to cover the plate overlay image, to extract the intersecting blood vessels point except mine lamp (or glasses) blocks position, and then obtain the crossover points distributed proper vector; Top n component to all crossover points distributed proper vectors sorts from small to large according to φ, obtains SV
i, i=1,2 ..., M;
(9) whether basis wears mine lamp (or glasses), the distribution characteristics vector SV after the intersecting blood vessels point sequence of the miner's to be identified that calculating is extracted thermal infrared images
i, i=1,2 ..., corresponding sequence crossover points distributed proper vector SW corresponding to image that wears or do not wear mine lamp, glasses of each miner in M and miner's face characteristic storehouse
jMatching degree r
Match, j=1 wherein, 2 ..., M ', M ' are the point of crossing sums that extracts in current miner's image to be compared in feature database, according to matching degree, obtain recognition result.
Claims (4)
1. thermal infrared images face identification system under a coal mine, is characterized in that, described system comprises processing unit, thermal infrared imager, mining temperature sensor, storage unit, communication line, aboveground server and aboveground attendance data storehouse; Described thermal infrared imager, be used for gathering people's face thermal infrared images; Described mining temperature sensor, be used for gathering environment temperature, with people's face thermal infrared images that described thermal infrared imager is collected, carries out temperature compensation; Described storage unit, what be used for that storage miner's face characteristic storehouse establishment stage obtains is wearing or the crossover points distributed proper vector extracted of the thermal infrared images under wearing spectacles, mine lamp condition not from all miners; Described communication line, be used for processing unit is passed to described aboveground server to the recognition result information of inputting people's face thermal infrared images to be identified; Described aboveground server is used for receiving and distributing the recognition result that the down-hole treatment unit comes by the communication line transmission; Described aboveground attendance data storehouse, for record discrepancy well miner's identity, temporal information.
2. thermal infrared images face identification system under a kind of coal mine according to claim 1, is characterized in that, described aboveground server comprises aboveground server end communication module, work attendance module and alarm module; Aboveground server end communication module is responsible for communicating by letter with the communication module of down-hole treatment unit, receives the recognition result to people's face to be identified; The work attendance module is used for recognition result is write described aboveground attendance data storehouse; Alarm module is used for warning aboveground managerial personnel.
3. thermal infrared images face identification system under a kind of coal mine according to claim 1, is characterized in that, described processing unit comprises control module, recognition of face and processing module, proper vector access module and communication module; Described recognition of face and processing module are used for obtaining people's face identity; Described proper vector access module is used for the crossover points distributed proper vector that access miner face characteristic storehouse establishment stage obtains; Described communication module is used for that described recognition of face is obtained result with the processing module processing and transfers to described aboveground server.
4. thermal infrared images face identification system under a kind of coal mine according to claim 3, is characterized in that, described control module comprises temperature acquisition control module and thermal infrared images controlling of sampling module; Described temperature acquisition control module is used for controlling described mining temperature sensor,, with certain sampling period collection environment temperature, with people's face thermal infrared images that described thermal infrared imager is collected, carries out temperature compensation; Described thermal infrared images controlling of sampling module is used for controlling described thermal infrared imager collection people face thermal infrared images.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013202745518U CN203299842U (en) | 2013-05-20 | 2013-05-20 | Thermal infrared image face recognition system adapted in underground coal mine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013202745518U CN203299842U (en) | 2013-05-20 | 2013-05-20 | Thermal infrared image face recognition system adapted in underground coal mine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN203299842U true CN203299842U (en) | 2013-11-20 |
Family
ID=49575840
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013202745518U Expired - Fee Related CN203299842U (en) | 2013-05-20 | 2013-05-20 | Thermal infrared image face recognition system adapted in underground coal mine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN203299842U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451543A (en) * | 2017-07-14 | 2017-12-08 | 广东欧珀移动通信有限公司 | Event-handling method and Related product |
| CN113139413A (en) * | 2020-08-07 | 2021-07-20 | 西安天和防务技术股份有限公司 | Personnel management method and device and electronic equipment |
-
2013
- 2013-05-20 CN CN2013202745518U patent/CN203299842U/en not_active Expired - Fee Related
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107451543A (en) * | 2017-07-14 | 2017-12-08 | 广东欧珀移动通信有限公司 | Event-handling method and Related product |
| CN113139413A (en) * | 2020-08-07 | 2021-07-20 | 西安天和防务技术股份有限公司 | Personnel management method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103246883B (en) | A kind of underground coal mine thermal infrared images face identification method | |
| CN104091176B (en) | Portrait comparison application technology in video | |
| CN106203294B (en) | The testimony of a witness based on face character analysis unifies auth method | |
| CN100481112C (en) | Human face recognition system and recognition method | |
| CN102521565B (en) | Garment identification method and system for low-resolution video | |
| CN105243386B (en) | Face living body judgment method and system | |
| CN109101871A (en) | A kind of living body detection device based on depth and Near Infrared Information, detection method and its application | |
| CN104965235A (en) | Security check system and method | |
| CN103902962B (en) | One kind is blocked or the adaptive face identification method of light source and device | |
| CN106203415A (en) | A kind of bank based on Digital Image Processing card number automatic identification equipment | |
| CN103984941A (en) | Face recognition checking-in method and device thereof | |
| CN101140620A (en) | Human face recognition system | |
| CN102497531A (en) | Human face recognition wireless video monitoring system based on cloud computing platform | |
| CN105590097A (en) | Security system and method for recognizing face in real time with cooperation of double cameras on dark condition | |
| Kim et al. | A survey and proposed framework on the soft biometrics technique for human identification in intelligent video surveillance system | |
| CN107609515B (en) | Double-verification face comparison system and method based on Feiteng platform | |
| CN105975938A (en) | Smart community manager service system with dynamic face identification function | |
| CN101739742A (en) | Networking type multi-channel access control and attendance system | |
| CN101587485A (en) | Face information automatic login method based on face recognition technology | |
| CN103996033B (en) | Human identification method based on tracking of human face five-sense-organ coordinates | |
| CN104751143A (en) | Person and credential verification system and method based on deep learning | |
| CN102521571A (en) | Multimode biological identifying device and method thereof | |
| CN107516076A (en) | Portrait identification method and device | |
| CN202815870U (en) | Certificate photograph and face automatic identification system | |
| CN205354146U (en) | Human dual feature recognition module |
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: 20131120 Termination date: 20200520 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |