CN108346265A - A kind of risk object identification alarm method and device - Google Patents
A kind of risk object identification alarm method and device Download PDFInfo
- Publication number
- CN108346265A CN108346265A CN201810070055.8A CN201810070055A CN108346265A CN 108346265 A CN108346265 A CN 108346265A CN 201810070055 A CN201810070055 A CN 201810070055A CN 108346265 A CN108346265 A CN 108346265A
- Authority
- CN
- China
- Prior art keywords
- module
- face
- image
- risk object
- alarm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
- G08B21/22—Status alarms responsive to presence or absence of persons
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
A kind of risk object identification alarm method and device, include the following steps, 1) public safety information collecting device the image collected is detected by data acquisition module;2) image processing module processing is carried out;3) face recognition processing module is used;4) recognition result of step 3) the recognition of face processing module is received;5) data and control command sent according to the step 4) controller carry out actuation of an alarm.The present invention can solve existing public safety hazards target difficulty and find, be not easy the technical issues of timely processing.
Description
Technical field
The invention belongs to technical field of face recognition, and in particular to a kind of risk object identification alarm method and device.
Background technology
With the development of society, people’s lives rhythm is constantly accelerated, social scope is gradually expanded, the crowd contacted and
The range of arrival is also increasingly wider.Corresponding to this, the security risk of people at one's side is also increasingly increasing, this is in the main block in city
The larger place of equal flows of the people is particularly evident.Existing risk object identifies alarm system, cannot timely and accurately determine danger
Target, and do not have target identification learning functionality, alarm operation can not be carried out automatically.It needs to install in application scenarios simultaneously
A large amount of system attachment devices, cause cost high, and the portability between different scenes substantially reduces.Therefore, one is established
Efficiently perfect risk object identification alarm system is very necessary.Ensure that people's safety of life and property is very heavy
It wants, people are intended in the Safety Cities public environment for oneself being in the crowd that is safe from danger, and enjoy good life, this is
Current demand there is an urgent need to.
Existing technology mainly has following deficiency:
One, due to cost, the reason of system complexity, now on the market in this this kind of product few in number, big portion
Subsystem composition is complicated, and operation and maintenance cost is larger;
Two, due to function is limited, a kind of common risk object identification alarm module is difficult on the market now
The place that flow of the people is larger, mobility is strong plays due effect;
Three, due to the deficiency of technology, existing product is difficult to carry out precise acquisition, precisely identification to object on the move;
Four, since wireless communication module is not advanced enough, existing product has that alarm is not prompt enough.
Application No. is 201610525468.1 Chinese patent, to disclose a kind of Civil Aviation Airport terminal safety long-distance real-time
Supervisory information system, it includes, including video data acquiring module, sensor assembly, data processing module, recognition of face mould
Block, dangerous play determination module, central processing unit, airport environment regulation and control module and three-dimensional simulation module, it has suspect's face
It identifies, the function of dangerous play decision-making function, it is dangerous convenient for finding in time, to make corresponding emergency measure, greatly improve
Safety in entire terminal.
Invention content
The main purpose of the present invention is to provide a kind of risk object identification alarm method and devices, can solve the prior art
Recognition efficiency is not high, and does not have the ability for being iterated study, and risk object finds not accurate enough technical problem.
In order to solve the above-mentioned technical problem, present invention employs following technical schemes:
A kind of risk object identification alarm method and device, it includes the following steps:
1) public safety information collecting device the image collected is detected by data acquisition module;
2) pass through image processing module to handle, step 1 the image collected is divided using training aids trained in advance
Class, determine it is described it is collected whether be facial image, obtain face image set;
3) face recognition processing module is used, to having previously been stored in the risk object personage people in recognition of face processing module
The face image set that image and step 2) in face head portrait library collect carries out face matching, and the dangerous mesh that matching is obtained
Target specifying information is packaged into note data packet;
4) recognition result of step 3) the recognition of face processing module is received, and alarm module is controlled according to recognition result
Carry out the keying of alarm module;Various control commands for receiving human-machine operation, and order these according to preset algorithm
It is sent to corresponding module;The reception of data for each module, and be sent to database and stored;
5) data and control command sent according to the step 4) controller carry out actuation of an alarm.
In step 1), following steps are specifically used:
1) in the main block in this city and key crossing setting information harvester, and therefrom by certain space-time rule
Extract the image information on a certain ground of certain time period;
2) all information of main block and key crossing are adopted as a result, according to when and where for above-mentioned steps
The image information of acquisition means acquisition forms image set by certain space-time rule.
In step 2), following steps are specifically used:
1) grader based on Haar input feature vector bases is used:To rectangular image area and progress thresholding;
2) integral image techniques accelerate the calculating of the value of 45 degree of rectangular image area rotations, this picture structure by with
To accelerate the calculating of class Haar input feature vectors;
3) the grader node of two classification problems (face with non-face) is created using Adaboost;
4) the composition screening type cascade of grader node, a node is a classifiers of Adaboost types, screening
Go out the face images in image set.
In step 3), following steps are specifically included:
1) algorithm is set, face key point is set, builds human face recognition model function;
2) on the basis of the above-described procedure to having previously been stored in risk object personage's face head in recognition of face processing module
As in library image and the obtained face image set of the step 2 carry out critical point detection, the extraction of description, the phase based on distance
It is calculated like property, obtains recognition result;
3) on the basis of above-mentioned steps, using machine learning algorithm statistics, analysis and identification as a result, and being carried out based on result
Learning training, if similarity reaches some threshold value, we if think gained facial image whether be risk object personage, otherwise
Not think be;
4) on the basis of above-mentioned steps, and the specifying information of risk object is packaged into note data packet.
In step 4), following steps are specifically included:
1) it is used to receive the note data packet of the step 3), the various control commands for receiving human-machine operation, and presses
These orders are sent to corresponding module according to preset algorithm;
2) keying that alarm module carries out alarm module is controlled on the basis of step 3);
3) database is sent to be stored.
In step 5), following steps are specifically included:
1) obtaining step 4) instruction to alarm module.
2) it if it is enabled instruction, then sends note data packet to " 12110 " and alarms.
Including information collecting device, controller module, target identification processing module, communication module, alarm module and power supply
Module, information collecting device are electronic eyes, and electronic eyes is mounted on the monitoring point of area to be monitored, the signal of information collecting device
First signal input part of output end and controller module connects, the signal output end and controller mould of target identification processing module
The second signal input terminal of block connects;The first signal output end of controller module connects the signal input part of alarm module, alarm
The signal output end of module is communicated by communication module with the police, and it is defeated that controller module power input end connects power module electric power
Outlet.
Above- mentioned information harvester is that external is taken pictures electronic eyes.
Above-mentioned training aids is Viola-Jones graders.
The method have the advantages that:
Compared with the prior art, the present invention is monitored the main block in city in a manner of a kind of low cost, high efficiency and with more
Good portability is faster more easily applied to different scenes, can effectively prevent the generation with dangerous situation;Using
Machine learning algorithm can constantly identify risk object to the ability of risk object identified and have self-teaching and improve
Precision and accuracy, in a certain range of space-time environment, risk object personage being identified from crowd in time, is gone forward side by side
Row alarm operation, effectively maintains the safety in monitoring range;The autonomous setting that can carry out risk object, to be searched
It does, the police is helped to build intelligent monitoring network, dramatically reduce manpower monitoring pressure, reduce police strength deployment spending, electricity consumption
The mode of son realizes the gridding supervision to society.
Description of the drawings
The invention will be further described with reference to the accompanying drawings and examples;
Fig. 1 is the flow chart of present invention identification alarm method;
Fig. 2 is the structural schematic diagram of present invention identification alarm module.
Specific implementation mode
A kind of risk object identification alarm method and device, it includes the following steps:
1) 1 the image collected of public safety information collecting device is detected by data acquisition module;
2) pass through image processing module to handle, step 1) the image collected is divided using training aids trained in advance
Class, determine it is described it is collected whether be facial image, obtain face image set;
3) face recognition processing module is used, to having previously been stored in the risk object personage people in recognition of face processing module
The face image set that image and step 2) in face head portrait library collect carries out face matching, and the dangerous mesh that matching is obtained
Target specifying information is packaged into note data packet;
4) recognition result of step 3) the recognition of face processing module is received, and alarm module is controlled according to recognition result
Carry out the keying of alarm module;Various control commands for receiving human-machine operation, and order these according to preset algorithm
It is sent to corresponding module;The reception of data for each module, and be sent to database and stored;
5) data and control command sent according to the step 4) controller carry out actuation of an alarm.
In step 1), following steps are specifically used:
1) in the main block in this city and key crossing setting information harvester 1, and by certain space-time rule from
The image information on the middle a certain ground for extracting certain time period;
2) all information of main block and key crossing are adopted as a result, according to when and where for above-mentioned steps
The image information that acquisition means 1 acquire forms image set by certain space-time rule.
In step 2), following steps are specifically used:
1) using the grader based on Haar input feature vector bases to rectangular image area and progress thresholding;
2) integral image techniques accelerate the calculating of the value of 45 degree of rectangular image area rotations, this picture structure by with
To accelerate the calculating of class Haar input feature vectors;
3) the grader node of two classification problems (face with non-face) is created using Adaboost;
4) the composition screening type cascade of grader node, a node is a classifiers of Adaboost types, screening
Go out the face images in the image set obtained in the step 1).
In step 3), following steps are specifically included:
1) algorithm is set, face key point is set, builds human face recognition model function;
2) on the basis of the above-described procedure to having previously been stored in risk object personage's face head in recognition of face processing module
As in library image and the obtained face image set of the step 2 carry out critical point detection, the extraction of description, the phase based on distance
It is calculated like property, obtains recognition result;
3) on the basis of above-mentioned steps, using machine learning algorithm statistics, analysis and identification as a result, and being carried out based on result
Learning training, if similarity reaches some threshold value, we if think gained facial image whether be risk object personage, otherwise
Not think be;
4) on the basis of above-mentioned steps, and the specifying information of risk object is packaged into note data packet.
For explaining in detail for step 3:
Face critical point detection positions human face five-sense-organ feature etc. based on caffe frames are improved, obtains needing to learn
Human face recognition model function:L=g (S, M), wherein S are the facial images of input, and M is the model parameter for needing to learn, and L is
Need the face coordinate position parameter detected, L ∈ [(x1,y1),(x2,y2),(x3,y3),(x4,y4)…,(xi,yi)], i=64,
Training convolutional neural networks return characteristic point coordinate.Traditional caffe frames do not support multi-tag, make its branch used here as hdf5
Hold multi-tag.
Description extracts, and (s is characterized a size degree for place layer to radius 8s border circular areas around selected characteristic vertex neighborhood
Amount), it calculates 45 ° of sectors and is finally selected in rotation sector at a certain angle with the haar wavelet characters of vertical direction in the horizontal direction
The direction of maximum value is selected as characteristic point direction.
In step 4), following steps are specifically included:
1) it is used to receive the note data packet of the step 3), the various control commands for receiving human-machine operation, and presses
These orders are sent to corresponding module according to preset algorithm;
2) keying that alarm module carries out alarm module is controlled on the basis of step 3);
3) database is sent to be stored.
In step 5), following steps are specifically included:
1) obtaining step 4) instruction to alarm module.
2) it if it is enabled instruction, then sends note data packet to " 12110 " and alarms.
Including information collecting device 1, controller module 2, target identification processing module 3, communication module 4,5 and of alarm module
Power module 6, it is characterised in that:Information collecting device 1 is electronic eyes, and electronic eyes is mounted on the monitoring point of area to be monitored,
The signal output end of described information harvester 1 is connect with the first signal input part of controller module 2, target identification processing mould
The signal output end of block 3 is connected with the second signal input terminal of controller module 2;2 first signal output end of controller module connects
The signal output end of the signal input part of module 5 taking alarm, alarm module 5 is communicated by communication module 4 with the police, controller mould
2 power input end of block connects 6 power output end of power module.
Described information harvester 1 is that external is taken pictures electronic eyes.
The training aids is Viola-Jones graders.
Optionally, controller module 2 is PIC18F87K22 controllers.
Optionally, target identification communication module 4 is NB-IoT wireless communication modules.
Optionally, alarm module 5 is CWT50105.
Wherein, used PIC18F87K22 series has used the high-performance of extremely low power dissipation (nanoWatt XLP) technology
MCU is integrated with 1Mb enhancements flash memory and 12,24 tunnels ADC, has a up to 16MIPS performances, 8x8 monocycle hardware multipliers,
Working frequency is up to 64MHz, operating voltage 1.8V-5.5V, there is 10 CCP/ECCP modules, 11 8/16 bit timing device/countings
Device module, two main synchronous serial interface modules have strict demand and using battery as the application of the energy especially suitable for power, it is ensured that
System timely responds to, and has ensured the safety and stability of hardware.
Wherein, use based on cellular narrowband Internet of Things (NB-IoT) is implemented in cellular network, and only consumption is about
The bandwidth of 180kHz can be deployed directly into GSM network etc., lower deployment cost be greatly reduced, and can realize smooth upgrade.The target
Identification communication module is NB-IoT wireless communication modules 4.
Wherein, used CWT5010 is the long-range I/O cell for having SMS function, its embedded real time operating system makes
With technical grade communication module and high-performance 32-bit processor, status input signal once changes, and can send short message immediately,
Ensure timely, the alarm module CWT5010 for sending information.
Claims (9)
1. a kind of risk object identification alarm method and device, which is characterized in that it includes the following steps:
1) public safety information collecting device (1) the image collected is detected by data acquisition module;
2) pass through image processing module to handle, be classified to step 1) the image collected using training aids trained in advance,
Determine it is described it is collected whether be facial image, obtain face image set;
3) face recognition processing module (3) is used, to having previously been stored in the risk object personage people in recognition of face processing module
The face image set that image and step 2) in face head portrait library collect carries out face matching, and the dangerous mesh that matching is obtained
Target specifying information is packaged into note data packet;
4) recognition result of step 3) the recognition of face processing module is received, and alarm module (5) is controlled according to recognition result
Carry out the keying of alarm module;Various control commands for receiving human-machine operation, and order these according to preset algorithm
It is sent to corresponding module;The reception of data for each module, and be sent to database and stored;
5) data and control command sent according to the step 4) controller carry out actuation of an alarm.
2. risk object identification alarm method according to claim 1 and device, which is characterized in that in step 1), tool
Body uses following steps:
1) in the main block in this city and key crossing setting information harvester (1), and therefrom by certain space-time rule
Extract the image information on a certain ground of certain time period;
2) all information collections of main block and key crossing are filled as a result, according to when and where for above-mentioned steps
The image information for setting (1) acquisition forms image set by certain space-time rule.
3. risk object identification alarm method according to claim 1 and device, which is characterized in that in step 2), tool
Body uses following steps:
1) grader based on Haar input feature vector bases is used:To rectangular image area and progress thresholding;
2) integral image techniques accelerate the calculating of the value of 45 degree of rotations of rectangular image area, this picture structure is used to add
The calculating of fast class Haar input feature vectors;
3) the grader node of two classification problems is created using Adaboost;
4) the composition screening type cascade of grader node, a node is a classifiers of Adaboost types, filters out figure
Face images in image set.
4. risk object identification alarm method according to claim 1 and device, which is characterized in that in step 3), tool
Body includes the following steps:
1) algorithm is set, face key point is set, builds human face recognition model function;
2) on the basis of the above-described procedure to having previously been stored in risk object personage's face head portrait library in recognition of face processing module
In the obtained face image set of image and the step 2) carry out critical point detection, the extraction of description, based on the similar of distance
Property calculate, obtain recognition result;
3) on the basis of above-mentioned steps, using machine learning algorithm statistics, analysis and identification as a result, and being learnt based on result
Training, if similarity reaches some threshold value, we if think gained facial image whether be risk object personage, otherwise it is assumed that
It is not;
4) on the basis of above-mentioned steps, and the specifying information of risk object is packaged into note data packet.
5. risk object identification alarm method according to claim 1 or 4 and device, which is characterized in that in step 4),
Specifically include following steps:
1) the note data packet for receiving the step 3), the various control commands for receiving human-machine operation, and according to pre-
If algorithm these orders are sent to corresponding module;
2) keying that alarm module carries out alarm module is controlled on the basis of step 3);
3) database is sent to be stored.
6. risk object identification alarm method according to claim 5 and device, which is characterized in that in step 5), tool
Body includes the following steps:
1) obtaining step 4) instruction to alarm module.
2) it if it is enabled instruction, then sends note data packet to " 12110 " and alarms.
7. risk object identification alarm method according to claim 1 and device, it is characterised in that:It is filled including information collection
(1), controller module (2), target identification processing module (3), communication module (4), alarm module (5) and power module (6) are set,
It is characterized in that:Information collecting device (1) is electronic eyes, and electronic eyes is mounted on the monitoring point of area to be monitored, described information
The signal output end of harvester (1) is connect with the first signal input part of controller module (2), target identification processing module
(3) signal output end is connected with the second signal input terminal of controller module (2);(2) first signal of controller module exports
The signal output end of the signal input part of end connection alarm module (5), alarm module (5) is logical by communication module (4) and the police
Letter, controller module (2) power input end connect power module (6) power output end.
8. risk object identification alarm method according to claim 7 and device, it is characterised in that:Described information acquisition dress
It is that external is taken pictures electronic eyes to set (1).
9. risk object identification alarm method according to claim 1 and device, it is characterised in that:The training aids is
Viola-Jones graders.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810070055.8A CN108346265B (en) | 2018-01-24 | 2018-01-24 | A kind of risk object identification alarm method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810070055.8A CN108346265B (en) | 2018-01-24 | 2018-01-24 | A kind of risk object identification alarm method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108346265A true CN108346265A (en) | 2018-07-31 |
CN108346265B CN108346265B (en) | 2019-02-22 |
Family
ID=62960917
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810070055.8A Active CN108346265B (en) | 2018-01-24 | 2018-01-24 | A kind of risk object identification alarm method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108346265B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109242747A (en) * | 2018-09-14 | 2019-01-18 | 海大恒网络科技(昆山)有限公司 | A kind of public safety anti-terrorism integrated application platform analysis system and method |
CN109447048A (en) * | 2018-12-25 | 2019-03-08 | 苏州闪驰数控系统集成有限公司 | A kind of artificial intelligence early warning system |
CN109508652A (en) * | 2018-10-25 | 2019-03-22 | 国影(北京)科技有限责任公司 | Viewing number statistical method, device and electronic equipment |
CN109597575A (en) * | 2018-11-29 | 2019-04-09 | 中国科学院合肥物质科学研究院 | A kind of storage of sectional type data and read method based on HDF5 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8552873B2 (en) * | 2010-12-28 | 2013-10-08 | Automotive Research & Testing Center | Method and system for detecting a driving state of a driver in a vehicle |
CN106446791A (en) * | 2016-08-31 | 2017-02-22 | 彭青 | Smart city public monitoring system |
CN206164722U (en) * | 2016-09-21 | 2017-05-10 | 深圳市泛海三江科技发展有限公司 | Discuss super electronic monitoring system based on face identification |
CN206807645U (en) * | 2016-03-30 | 2017-12-26 | 冯华飞 | A kind of monitor based on recognition of face |
-
2018
- 2018-01-24 CN CN201810070055.8A patent/CN108346265B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8552873B2 (en) * | 2010-12-28 | 2013-10-08 | Automotive Research & Testing Center | Method and system for detecting a driving state of a driver in a vehicle |
CN206807645U (en) * | 2016-03-30 | 2017-12-26 | 冯华飞 | A kind of monitor based on recognition of face |
CN106446791A (en) * | 2016-08-31 | 2017-02-22 | 彭青 | Smart city public monitoring system |
CN206164722U (en) * | 2016-09-21 | 2017-05-10 | 深圳市泛海三江科技发展有限公司 | Discuss super electronic monitoring system based on face identification |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109242747A (en) * | 2018-09-14 | 2019-01-18 | 海大恒网络科技(昆山)有限公司 | A kind of public safety anti-terrorism integrated application platform analysis system and method |
CN109508652A (en) * | 2018-10-25 | 2019-03-22 | 国影(北京)科技有限责任公司 | Viewing number statistical method, device and electronic equipment |
CN109597575A (en) * | 2018-11-29 | 2019-04-09 | 中国科学院合肥物质科学研究院 | A kind of storage of sectional type data and read method based on HDF5 |
CN109597575B (en) * | 2018-11-29 | 2021-09-28 | 中国科学院合肥物质科学研究院 | HDF 5-based block-type data storage and reading method |
CN109447048A (en) * | 2018-12-25 | 2019-03-08 | 苏州闪驰数控系统集成有限公司 | A kind of artificial intelligence early warning system |
CN109447048B (en) * | 2018-12-25 | 2020-12-25 | 苏州闪驰数控系统集成有限公司 | Artificial intelligence early warning system |
Also Published As
Publication number | Publication date |
---|---|
CN108346265B (en) | 2019-02-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108346265B (en) | A kind of risk object identification alarm method | |
Liu et al. | Drone detection based on an audio-assisted camera array | |
CN105913559B (en) | A kind of ATM in bank intelligent control method based on body-sensing technology | |
CN103325125B (en) | Based on the motion target tracking method improving multi-instance learning algorithm | |
US11276185B2 (en) | Target tracking method and apparatus | |
Liu et al. | Measuring indoor occupancy in intelligent buildings using the fusion of vision sensors | |
CN104298353A (en) | Inverse kinematics based vehicle monitoring and burglary preventing method and system | |
Vaddi | Efficient object detection model for real-time UAV applications | |
CN103824070A (en) | Rapid pedestrian detection method based on computer vision | |
CN112270267B (en) | Camera shooting identification system capable of automatically capturing line faults | |
EP3723046A1 (en) | Target tracking method and device | |
CN204679863U (en) | Based on the intelligent substation inspection system of many rotor wing unmanned aerial vehicles | |
CN102982341A (en) | Self-intended crowd density estimation method for camera capable of straddling | |
CN103105924B (en) | Man-machine interaction method and device | |
TW201911118A (en) | Method for tracing track of target in cross regions, and data processing method, apparatus and system | |
CN105022999A (en) | Man code company real-time acquisition system | |
CN108748184B (en) | Robot patrol method based on regional map identification and robot equipment | |
CN202940921U (en) | Real-time monitoring system based on face identification | |
CN105404867A (en) | Vision-based substation disconnecting switch status recognition method | |
Shen et al. | Person tracking and frontal face capture with uav | |
CN106296743A (en) | A kind of adaptive motion method for tracking target and unmanned plane follow the tracks of system | |
CN104159088B (en) | A kind of long-distance intelligent vehicle monitoring system and method | |
CN110457987A (en) | Face identification method based on unmanned plane | |
WO2023211499A1 (en) | Machine learning real property object detection and analysis apparatus, system, and method | |
CN201130382Y (en) | Multifunctional identification automatic recognition robot |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20210728 Address after: 443002 No.73 zhenjingshan Road, Xiling District, Yichang City, Hubei Province Patentee after: Gezhouba Communication Technology Co.,Ltd. Address before: 443002 No. 8, University Road, Yichang, Hubei Patentee before: CHINA THREE GORGES University |