CN110163034A - A kind of listed method of aircraft surface positioning extracted based on optimal characteristics - Google Patents
A kind of listed method of aircraft surface positioning extracted based on optimal characteristics Download PDFInfo
- Publication number
- CN110163034A CN110163034A CN201810162180.1A CN201810162180A CN110163034A CN 110163034 A CN110163034 A CN 110163034A CN 201810162180 A CN201810162180 A CN 201810162180A CN 110163034 A CN110163034 A CN 110163034A
- Authority
- CN
- China
- Prior art keywords
- aircraft
- optimal
- sample
- dimension
- follows
- 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
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000013528 artificial neural network Methods 0.000 claims abstract description 4
- 230000006870 function Effects 0.000 claims description 28
- 239000011159 matrix material Substances 0.000 claims description 18
- 238000000605 extraction Methods 0.000 claims description 13
- 210000002569 neuron Anatomy 0.000 claims description 12
- 239000013598 vector Substances 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 238000004422 calculation algorithm Methods 0.000 claims description 3
- 238000009434 installation Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 241000256259 Noctuidae Species 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000003707 image sharpening Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of aircraft surfaces extracted based on optimal characteristics to position listed method, comprising: the industrial camera for acquiring number of registration under the aircraft wing buries dress layout；Acquired image sample set is analyzed, the best identified dimension collection C of sample mode is calculatedvb；The fitness for calculating best identified dimension collection, determines best identified dimension collection；According to determining best identified dimension, number of registration under the wing of aircraft is identified using BP neural network, realizes that accurately positioning is listed.The aircraft underwing that the industrial camera that the present invention can be detached from taxiway according to runway takes divides photo, quickly recognizes the number of registration of aircraft, to realize the terrestrial positioning of aircraft and be listed.
Description
Technical Field
The invention belongs to the field of computer communication, pattern recognition and aviation safety, and particularly relates to an aircraft ground positioning and listing method based on optimal feature extraction.
Background
The global aviation industry generates many unsafe events of aircraft sliding conflict every year, scratch occurs in the light case, and casualties occur in the heavy case. Therefore, various aircraft positioning and registering technologies have been developed, that is, the ground aircraft is positioned by using the field monitoring radar, and the calling number and the registration number are hung for the aircraft by using the information interface. The existing positioning and listing technology generally adopts a system fusion technology, one technology is that an aircraft is positioned by an air traffic control radar and a field monitoring radar, and then positioning and listing are realized by aircraft information of an air traffic control automation system; the tag hanging of the technology completely depends on aircraft information of an air traffic control automation system, and if the flight correction report of the air traffic control automation is not received timely, tag hanging errors can be caused, and sliding conflicts can be caused more easily. The other method is based on ADS-B technology, but the ADS-B system is often turned off during the landing taxiing stage of the aircraft, and positioning cannot be achieved.
Disclosure of Invention
In order to solve the problems, the invention provides an aircraft ground positioning and listing method based on optimal feature extraction, which can realize accurate aircraft ground positioning and listing without depending on airborne equipment.
The technical scheme adopted by the invention is as follows:
an aircraft ground positioning and listing method based on optimal feature extraction is characterized by comprising the following steps: the method comprises the following steps: A. the embedded layout of the industrial camera is used for collecting the registration number under the wing of the aircraft; B. analyzing the acquired image sample set, and calculating the optimal identification dimension set C of the sample modevb(ii) a C. Calculating the fitness of the optimal identification dimension set, and determining the optimal identification dimension set; D. and identifying the registration number under the wing of the aircraft by adopting a BP (back propagation) neural network according to the determined optimal identification dimension, so as to realize accurate positioning and listing.
Preferably, the specific implementation manner of step a is as follows:
(1) the lens is specifically mounted as follows:
suppose that the span of the aircraft is SPwThe maximum wingspan of the airplane type which can be taken off and landed and is determined according to the grade of a flight area of an airport is Max (SP)w) Minimum wingspan Min (SP)w) The maximum airplane type airfoil surface ground clearance is HwThe maximum offset distance of the aircraft during taxiing on the road surface is dmax：
① the lenses are mounted in pairs and are equidistantly distributed on two sides of the central line of the road surface, and the connecting line of the lenses of the same pair is vertical to the central line of the road surface;
② vertical distance of lens from the center line of road surfaceSeparation device
③ Angle of lens VaThe determination method comprises the following steps:
thus, there is a maximum viewing angle Max (V)a) Comprises the following steps:
according to the steps, the type selection and installation parameters of the camera lens can be determined;
(2) and then determining the distribution of the industrial cameras in the whole flight area, and ensuring that the aircraft slides into any station area to pass through at least one group of lenses no matter what direction the runway is used:
the cameras are distributed at two ends of the center line of the following road surface:
① all quick exit lanes;
② all running-sliding contact tracks;
③ all the slippery-slippery communication tracks;
④ all running-sliding end taxiways;
as described above, the task of image acquisition of the registration number under the wing of the aircraft from landing to any station ground can be reliably completed.
Preferably, the specific implementation manner of step B is:
the method for calculating the optimal recognition dimension set of the sample pattern is as follows:
assuming that the characteristic dimension obtained after sample preprocessing is n, t types exist after sample clustering,sample internal divergence matrix is MinThe inter-class divergence matrix is MoutThe global divergence matrix is MaIntroducing a column vector CvThe fisher identification function of the system is made as:
best recognition dimension set CvbThe following conditions should be satisfied: sample set in CvbProjection in the direction will cause the class to have the smallest internal divergence matrix MinAnd the maximum inter-class divergence matrix MoutI.e. the set of optimal recognition dimensions CvbIs the generalized linear characteristic equation: mout·Cvb＝λ·Min·CvbThe eigenvector corresponding to the maximum eigenvalue;
solving p optimal identification dimensions according to the formulaThen the (p +1) th best recognition dimensionThe following can be calculated:to obtain and makeObtaining a maximum value;
in order to make the samples have better separability, the algorithm is extended, and to be regarded as a whole and,the solution is taken as another whole, so that:
the (p +1) th best recognition dimension can be directly obtained by the above formula
Because the collection of the samples is independent, the samples meet the statistic non-correlation, and the optimal recognition transformation theorem according to the statistic non-correlation comprises the following steps:
wherein
I is an identity matrix;
the (p +1) th optimal identification dimension formula of the independent sample obtained according to the formula (4) obtained by the extended theory is as follows:
wherein,
according to equations (5) and (7), the optimal recognition dimension set C of the sample pattern can be calculatedvb。
Preferably, the specific implementation manner of step C is:
introducing a fitness function to describe the fitness of the obtained optimal identification dimension, wherein the fitness function is as follows:
assuming the optimal recognition dimension set obtained in the current stepWhere m is Min (n, t),
fitness function
f(Cvb) E (0,1), and the larger the value is, the better, and the fitness of the current optimal identification dimension can be intuitively understood according to the fitness function value.
Preferably, the specific implementation manner of step D is:
(1) dividing a training set and a test set, and setting network parameters:
let the connection weight of input layer-hidden layer be wihThe hidden layer-output layer connection weight is whoWith hidden layer neuron threshold ofOutput layer neuron threshold ofInput samples are recorded asThe output vector of the output layer is notedThe number of training samples is N-1, 2, …, N, and the error function adopts root mean square function;
(2) computing output of hidden layer neurons
Wherein,in order to make the input of the hidden layer,is an activation function;
(3) computing output vectors for an output layer
(4) Calculating the error e of the network:
wherein d ismRepresenting a desired output of the input sample;
(5) respectively calculating partial derivatives of the network error function on neurons of the hidden layer and the output layer:
(6) and feeding the error back to the network, and correcting the connection weight and the threshold of each layer:
(7) calculating a global network error E, and comparing to meet the tolerance error so as to decide whether to continue learning:
the invention has the beneficial effects that:
the method does not depend on airborne equipment of the aircraft, realizes ground positioning and registration of the aircraft, can accurately complete registration and registration of calling numbers for the aircraft, can early warn ground sliding conflict events of the aircraft, and has the advantages of low cost, simplicity and convenience in construction, stability and reliability.
Drawings
FIG. 1 is a schematic diagram of maximum viewing angle calculation;
fig. 2 is an overall flowchart of an aircraft ground positioning and listing method based on optimal feature extraction.
Detailed description of the invention
The invention relates to an aircraft ground positioning and listing method based on optimal feature extraction, which comprises the following steps:
A. the buried industrial camera is installed at a two-way sliding crossing of a runway, and in the layout, the aircraft is ensured to approach and fall to the ground, and the industrial camera lens is required to pass through any route in the process of separating from the runway until sliding into a parking position.
(1) The lens is specifically mounted as follows:
suppose that the span of the aircraft is SPwThe maximum wingspan of the airplane type which can be taken off and landed and is determined according to the grade of a flight area of an airport is Max (SP)w) Minimum wingspan Min (SP)w) The maximum airplane type airfoil surface ground clearance is HwThe maximum offset distance of the aircraft during taxiing on the road surface is dmax：
① the lenses are mounted in pairs and are equidistantly distributed on two sides of the central line of the road surface, and the connecting line of the lenses of the same pair is vertical to the central line of the road surface;
② vertical distance of lens from center line of road surface
③ Angle of lens VaThe determination method, as shown in fig. 1:
thus, there is a maximum viewing angle Max (V)a) The formula is as follows:
according to the steps, the type selection and installation parameters of the camera lens can be determined.
(2) The distribution of the industrial cameras throughout the flight area is then determined to ensure that the aircraft slides into any of the station areas past at least one set of shots regardless of the direction of runway used by the aircraft.
The cameras are distributed at two ends of the center line of the following road surface:
① all quick exit lanes;
② all running-sliding contact tracks;
③ all the slippery-slippery communication tracks;
④ all running-sliding end taxiways.
As described above, the task of image acquisition of the registration number under the wing of the aircraft from landing to any station ground can be reliably completed.
B. Analyzing the acquired image sample set, and calculating the optimal identification dimension set C of the sample modevb。
After graying, binary processing, image sharpening, noise reduction, word domain segmentation, normalization and feature extraction are carried out on the sample set, the image preprocessing flow is completed. However, in the image recognition process, the feature dimensionality is not as high as possible, and in the interval with the lower feature dimensionality, the recognition result accuracy is increased by increasing the dimensionality, and after a certain dimensionality is exceeded, the recognition accuracy is in a trend of obviously decreasing along with the increase of the feature dimensionality. Therefore, it is necessary to calculate the optimal recognition dimension set of the sample pattern, as follows:
assuming that the characteristic dimension obtained after sample preprocessing is n, t types exist after sample clustering, and the divergence matrix inside the sample is MinThe inter-class divergence matrix is MoutThe global divergence matrix is Ma. Introducing a column vector CvThe fisher identification function of the system is made as:
best recognition dimension set CvbThe following conditions should be satisfied: sample set in CvbProjection in the direction will cause the class to have the smallest internal divergence matrix MinAnd the maximum inter-class divergence matrix Mout. I.e. the set of best recognition dimensions CvbIs the generalized linear characteristic equation: mout·Cvb＝λ·Min·CvbAnd the feature vector corresponding to the maximum feature value.
Solving p optimal identification dimensions according to the formulaThen the (p +1) th best recognition dimensionThe following can be calculated:to obtain and makeThe maximum value is taken.
In order to make the samples have better separability, the algorithm is extended, and to be regarded as a whole and,the solution is taken as another whole, so that the following formula is given:
the (p +1) th best recognition dimension can be directly obtained by the above formula
In the embodiment of the present invention, the samples are collected independently, so that statistical non-correlation is satisfied between the samples, and the optimal recognition transformation theorem according to the statistical non-correlation includes:
wherein
I is an identity matrix;
the (p +1) th optimal identification dimension formula of the independent sample obtained according to the formula (4) obtained by the extended theory is as follows:
wherein,
according to equations (5) and (7), the optimal recognition dimension set C of the sample pattern can be calculatedvb。
C. And calculating the fitness of the optimal identification dimension set and determining the optimal identification dimension set.
From step B, the Fisher recognition function is knownThe function value of (2) expresses the size of separability of the optimal identification dimension, which is obtained by equation (7)So that F (C)v) The value is very small or close to 0, so that it is not necessary to change the valueSupplementing to the best recognition dimension set CvbIn (1).
Based on the above calculation, a fitness function is introduced to describe the fitness of the obtained optimal identification dimension, which is as follows:
assuming the optimal recognition dimension set obtained in the current stepWherein m is Min (m ═ Min: (m:)n,t)。
Fitness function
f(Cvb) E (0,1), and the larger the value, the better. And according to the fitness function value, the fitness of the currently-obtained optimal identification dimension can be intuitively known.
D. According to the determined optimal identification dimension, a machine learning method is applied, the BP neural network is adopted in the embodiment to identify the registration number under the wing of the aircraft, and finally accurate positioning and listing are achieved.
(1) Dividing a training set and a test set, and setting network parameters:
let the connection weight of input layer-hidden layer be wihThe hidden layer-output layer connection weight is whoWith hidden layer neuron threshold ofOutput layer neuron threshold ofInput samples are recorded asThe output vector of the output layer is notedThe number of training samples is N-1, 2, …, N, and the error function is root mean square function.
(2) Computing output of hidden layer neurons
Wherein,in order to make the input of the hidden layer,is an activation function.
(3) Computing output vectors for an output layer
(4) Calculating the error e of the network:
wherein d ismRepresenting the desired output of the input sample.
(5) Respectively calculating partial derivatives of the network error function on neurons of the hidden layer and the output layer:
(6) and feeding the error back to the network, and correcting the connection weight and the threshold of each layer:
(7) calculating a global network error E, and comparing to meet the tolerance error so as to decide whether to continue learning:
in conclusion, the aircraft ground positioning and listing method based on the optimal feature extraction according to the present invention is completed, and the overall flow chart is shown in fig. 2. The method does not depend on airborne equipment of the aircraft, realizes ground positioning and registration of the aircraft, can accurately complete registration and registration of calling numbers for the aircraft, can early warn ground sliding conflict events of the aircraft, and has the advantages of low cost, simplicity and convenience in construction, stability and reliability.
Claims (5)
1. An aircraft ground positioning and listing method based on optimal feature extraction is characterized by comprising the following steps: the method comprises the following steps: A. the embedded layout of the industrial camera is used for collecting the registration number under the wing of the aircraft; B. analyzing the acquired image sample set, and calculating the optimal identification dimension set C of the sample modevb(ii) a C. Calculating the fitness of the optimal identification dimension set, and determining the optimal identification dimension set; D. and identifying the registration number under the wing of the aircraft by adopting a BP (back propagation) neural network according to the determined optimal identification dimension, so as to realize accurate positioning and listing.
2. The aircraft ground positioning and listing method based on optimal feature extraction as claimed in claim 1, wherein: the specific implementation manner of the step A is as follows:
(1) the lens is specifically mounted as follows:
suppose that the span of the aircraft is SPwThe maximum wingspan of the airplane type which can be taken off and landed and is determined according to the grade of a flight area of an airport is Max (SP)w) Minimum wingspan Min (SP)w) The maximum airplane type airfoil surface ground clearance is HwThe maximum offset distance of the aircraft during taxiing on the road surface is dmax：
① the lenses are mounted in pairs and are equidistantly distributed on two sides of the central line of the road surface, and the connecting line of the lenses of the same pair is vertical to the central line of the road surface;
② vertical distance of lens from center line of road surface
③ Angle of lens VaThe determination method comprises the following steps:
thus, there is a maximum viewing angle Max (V)a) Comprises the following steps:
according to the steps, the type selection and installation parameters of the camera lens can be determined;
(2) and then determining the distribution of the industrial cameras in the whole flight area, and ensuring that the aircraft slides into any station area to pass through at least one group of lenses no matter what direction the runway is used:
the cameras are distributed at two ends of the center line of the following road surface:
① all quick exit lanes;
② all running-sliding contact tracks;
③ all the slippery-slippery communication tracks;
④ all running-sliding end taxiways;
as described above, the task of image acquisition of the registration number under the wing of the aircraft from landing to any station ground can be reliably completed.
3. The aircraft ground positioning and listing method based on optimal feature extraction as claimed in claim 2, wherein: the specific implementation manner of the step B is as follows:
the method for calculating the optimal recognition dimension set of the sample pattern is as follows:
assuming that the characteristic dimension obtained after sample preprocessing is n, t types exist after sample clustering, and the divergence matrix inside the sample is MinThe inter-class divergence matrix is MoutThe global divergence matrix is MaIntroducing a column vector CvThe fisher identification function of the system is made as:
best recognition dimension set CvbThe following conditions should be satisfied: sample set in CvbProjection in the direction will cause the class to have the smallest internal divergence matrix MinAnd the maximum inter-class divergence matrix MoutI.e. the set of optimal recognition dimensions CvbIs the generalized linear characteristic equation: mout·Cvb＝λ·Min·CvbThe eigenvector corresponding to the maximum eigenvalue;
solving p optimal identification dimensions according to the formulaThen the (p +1) th best recognition dimensionThe following can be calculated:to obtain and makeObtaining a maximum value;
in order to make the samples have better separability, the algorithm is extended, and to be regarded as a whole and,the solution is taken as another whole, so that:
the (p +1) th best recognition dimension can be directly obtained by the above formula
Because the collection of the samples is independent, the samples meet the statistic non-correlation, and the optimal recognition transformation theorem according to the statistic non-correlation comprises the following steps:
whereinI is an identity matrix;
the (p +1) th optimal identification dimension formula of the independent sample obtained according to the formula (4) obtained by the extended theory is as follows:
wherein,
according to equations (5) and (7), the optimal recognition dimension set C of the sample pattern can be calculatedvb。
4. The aircraft ground positioning and listing method based on optimal feature extraction as claimed in claim 3, wherein: the specific implementation manner of the step C is as follows:
introducing a fitness function to describe the fitness of the obtained optimal identification dimension, wherein the fitness function is as follows:
assuming the optimal recognition dimension set obtained in the current stepWhere m is Min (n, t),
fitness function
f(Cvb) E (0,1), and the larger the value is, the better, and the fitness of the current optimal identification dimension can be intuitively understood according to the fitness function value.
5. The aircraft ground positioning and listing method based on optimal feature extraction as claimed in claim 4, wherein: the specific implementation manner of the step D is as follows:
(1) dividing a training set and a test set, and setting network parameters:
let the connection weight of input layer-hidden layer be wihThe hidden layer-output layer connection weight is whoWith hidden layer neuron threshold ofOutput layer neuron threshold ofInput samples are recorded asThe output vector of the output layer is notedThe number of training samples is N-1, 2, …, N, and the error function adopts root mean square function;
(2) computing output of hidden layer neurons
Wherein,in order to make the input of the hidden layer,is an activation function;
(3) computing output vectors for an output layer
(4) Calculating the error e of the network:
wherein d ismRepresenting a desired output of the input sample;
(5) respectively calculating partial derivatives of the network error function on neurons of the hidden layer and the output layer:
(6) and feeding the error back to the network, and correcting the connection weight and the threshold of each layer:
(7) calculating a global network error E, and comparing to meet the tolerance error so as to decide whether to continue learning:
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110806322.5A CN113505714A (en) | 2018-02-27 | 2018-02-27 | Aircraft ground positioning and listing method based on optimal feature extraction |
| CN201810162180.1A CN110163034B (en) | 2018-02-27 | 2018-02-27 | Aircraft ground positioning and listing method based on optimal feature extraction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810162180.1A CN110163034B (en) | 2018-02-27 | 2018-02-27 | Aircraft ground positioning and listing method based on optimal feature extraction |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110806322.5A Division CN113505714A (en) | 2018-02-27 | 2018-02-27 | Aircraft ground positioning and listing method based on optimal feature extraction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110163034A true CN110163034A (en) | 2019-08-23 |
| CN110163034B CN110163034B (en) | 2021-07-23 |
Family
ID=67635738
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110806322.5A Withdrawn CN113505714A (en) | 2018-02-27 | 2018-02-27 | Aircraft ground positioning and listing method based on optimal feature extraction |
| CN201810162180.1A Active CN110163034B (en) | 2018-02-27 | 2018-02-27 | Aircraft ground positioning and listing method based on optimal feature extraction |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110806322.5A Withdrawn CN113505714A (en) | 2018-02-27 | 2018-02-27 | Aircraft ground positioning and listing method based on optimal feature extraction |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN113505714A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111210457A (en) * | 2020-01-08 | 2020-05-29 | 北京天睿空间科技股份有限公司 | Aircraft listing method combining video analysis and positioning information |
| WO2021109457A1 (en) * | 2019-12-06 | 2021-06-10 | 南京莱斯电子设备有限公司 | Airport airplane automatic labeling method based on self-learning policy |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101650944A (en) * | 2009-09-17 | 2010-02-17 | 浙江工业大学 | Method for distinguishing speakers based on protective kernel Fisher distinguishing method |
| CN103108170A (en) * | 2013-02-28 | 2013-05-15 | 深圳市云宙多媒体技术有限公司 | Video monitoring cloud platform controlling method and device |
| CN105868796A (en) * | 2016-04-26 | 2016-08-17 | 中国石油大学(华东) | Design method for linear discrimination of sparse representation classifier based on nuclear space |
| CN107273926A (en) * | 2017-06-12 | 2017-10-20 | 大连海事大学 | A kind of linear discriminant analysis dimension reduction method weighted based on cosine similarity |
-
2018
- 2018-02-27 CN CN202110806322.5A patent/CN113505714A/en not_active Withdrawn
- 2018-02-27 CN CN201810162180.1A patent/CN110163034B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101650944A (en) * | 2009-09-17 | 2010-02-17 | 浙江工业大学 | Method for distinguishing speakers based on protective kernel Fisher distinguishing method |
| CN103108170A (en) * | 2013-02-28 | 2013-05-15 | 深圳市云宙多媒体技术有限公司 | Video monitoring cloud platform controlling method and device |
| CN105868796A (en) * | 2016-04-26 | 2016-08-17 | 中国石油大学(华东) | Design method for linear discrimination of sparse representation classifier based on nuclear space |
| CN107273926A (en) * | 2017-06-12 | 2017-10-20 | 大连海事大学 | A kind of linear discriminant analysis dimension reduction method weighted based on cosine similarity |
Non-Patent Citations (5)
| Title |
|---|
| HONGYING ZHANG ET AL.: "RESEARCH ON CIVIL AVIATION AIRCRAFT REGISTRATION NUMBER RECOGNITION", 《2012 IEEE 2ND INTERNATIONAL CONFERENCE ON CLOUD COMPUTING AND INTELLIGENCE SYSTEMS》 * |
| SUPHOEBE: "神经网络学习 之 BP神经网络", 《CSDN》 * |
| 周丹等: "基于改进量子进化算法的特征选择", 《计算机工程与应用》 * |
| 王娜: "一种基于BP神经网络的车牌字符识别算法", 《电脑知识与技术》 * |
| 郭娟等: "基于加权Fisher统计不相关鉴别分析的人脸识别", 《计算机技术与应用进展》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021109457A1 (en) * | 2019-12-06 | 2021-06-10 | 南京莱斯电子设备有限公司 | Airport airplane automatic labeling method based on self-learning policy |
| CN111210457A (en) * | 2020-01-08 | 2020-05-29 | 北京天睿空间科技股份有限公司 | Aircraft listing method combining video analysis and positioning information |
| CN111210457B (en) * | 2020-01-08 | 2021-04-13 | 北京天睿空间科技股份有限公司 | Aircraft listing method combining video analysis and positioning information |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113505714A (en) | 2021-10-15 |
| CN110163034B (en) | 2021-07-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Al-qaness et al. | An improved YOLO-based road traffic monitoring system | |
| CN109191922B (en) | Large-scale four-dimensional track dynamic prediction method and device | |
| CN106127201A (en) | A kind of unmanned plane landing method of view-based access control model positioning landing end | |
| CN103971091B (en) | Automatic plane number recognition method | |
| CN104484870B (en) | Verify Plane location method | |
| Feng et al. | Mixed road user trajectory extraction from moving aerial videos based on convolution neural network detection | |
| CN112257151A (en) | Aircraft flight staging identification system | |
| CN110163034B (en) | Aircraft ground positioning and listing method based on optimal feature extraction | |
| CN111859247A (en) | Unmanned aerial vehicle operation risk assessment method based on satellite-based ADS-B data | |
| CN113869379A (en) | Data-driven aircraft energy anomaly identification method | |
| Minghui et al. | Deep learning enabled localization for UAV autolanding | |
| Kuchkorov et al. | Traffic and road sign recognition using deep convolutional neural network | |
| CN102567093A (en) | Berth type recognizing method applied in visual berth automatic guiding system | |
| Farhadmanesh et al. | General Aviation Aircraft Identification at Non-Towered Airports Using a Two-Step Computer Vision-Based Approach | |
| Martınez et al. | Forecasting unstable approaches with boosting frameworks and lstm networks | |
| Mirmohammadsadeghi et al. | Enhancements to the runway capacity simulation model using the asde-x data for estimating airports throughput under various wake separation systems | |
| CN106128170A (en) | A kind of aircraft berth intelligent indicating system | |
| Zhang et al. | Research on conflict detection model for taxi‐in process on the apron based on aircraft wingtip keypoint detection | |
| US20230267753A1 (en) | Learning based system and method for visual docking guidance to detect new approaching aircraft types | |
| CN117152706A (en) | Aircraft runway accumulated water identification method, device and system | |
| Farhadmanesh et al. | Implementing Haar Cascade Classifiers for Automated Rapid Detection of Light Aircraft at Local Airports | |
| Mirmohammadsadeghi et al. | Prediction of runway occupancy time and runway exit distance with feedforward neural networks | |
| Dhulipudi et al. | Geospatial object detection using machine learning-aviation case study | |
| WO2022261658A1 (en) | Aircraft identification | |
| CN114036640A (en) | Aircraft attitude determination system and aircraft |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210630 Address after: Room 213, Yijing building, No.1 Hengshan Road, Yantai Economic and Technological Development Zone, Shandong Province Applicant after: Shandong Yanhuang Industrial Design Co.,Ltd. Address before: 265400 94 Lengjia Zhuangzi Village, Wenquan sub district office, Zhaoyuan City, Yantai City, Shandong Province Applicant before: Leng Shuang |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |