CN108896972A - A kind of radar image simulation method based on image recognition - Google Patents
A kind of radar image simulation method based on image recognition Download PDFInfo
- Publication number
- CN108896972A CN108896972A CN201810651567.3A CN201810651567A CN108896972A CN 108896972 A CN108896972 A CN 108896972A CN 201810651567 A CN201810651567 A CN 201810651567A CN 108896972 A CN108896972 A CN 108896972A
- Authority
- CN
- China
- Prior art keywords
- image
- radar
- matrix
- picture
- eigenmatrix
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/40—Means for monitoring or calibrating
- G01S7/4052—Means for monitoring or calibrating by simulation of echoes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Abstract
A kind of radar image simulation method based on image recognition, contain main control unit and signal receiving unit, the main control unit is an industrial personal computer, main control unit built-in signal receiving unit and simulation software, and the simulation software includes having image display and picture recognition module；Picture recognition module has and the matched character matrix of radar image；Image display extracts the angle and color information of radar image from the digital information that radar is passed back, generate the true picture part in radar image, and the supplemental image part in radar image is generated with the interpolation parameter of character matrix selection, finally by two parts image superposition, to generate final high emulation radar image, it is shown on the display screen of main control unit.
Description
Technical field
The present invention relates to aircraft airborne radar image processing techniques, the specifically image simulation of aircraft airborne radar, especially
It is related in a kind of for the radar image simulation method based on image recognition without mating ground system.
Background technique
Radar is as airborne equipment, in the production and maintenance of aircraft, needs to be tested for the property radar, after qualification
It can install.
In the performance test of radar, some tests need to watch radar image just and can be carried out the judgement of test result.And
In production reality, many radars do not have dedicated display.The display of image depends on the display screen of other airborne equipments.Institute
Many times to need the performance test of radar and the performance test of other equipment to be placed on one in the performance test of radar
It rises and carries out.And in the actual work, it often will appear the case where lacking some equipment to which radar test can not be carried out, thus
Rhythm of production has been dragged slowly to cause economic loss.Since radar is expensive, Aircraft Production producer is past in order to accelerate rhythm of production
The test macro that can individually test radar is individually bought or manufactured toward meeting.
Currently, it can be typically necessary specially under no corollary equipment individually for the test macro that radar is tested for the property
It is mating customized, it is expensive.The radar image that the test macro is finally seen on a display screen is turned over by the signal that radar is passed back
It the straight line image that is translated on display screen and then is rushed between straight line image shown by actual signal according to certain interpolation coefficient benefit
Blank image, will both true picture part and supplemental image part superposition after form image.Its supplement figure of different radars
The interpolation coefficient and calculation method of picture are all different and underground, due to not knowing how the image algorithm of radar spreads out of radar
Signal on a display screen at figure, finally formed image often shows the image difference on screen farther out with airborne equipment, from
And the dependence test resultant error for resulting in the need for observation radar image is larger.
It is domestic at present to simulate the radar image shown on airborne indicator in computer terminal in unknown images algorithm,
It generally will use several mainstream algorithms to be attempted and which is observed to be closer to apart from real image, or directly using some calculation
Method imaging carries out error compensation, such basic blank of radar image simulation method based on image recognition to test result.
Summary of the invention
The purpose of the present invention is to provide a kind of radar image simulation methods based on image recognition, are existed with realizing to radar
When its unknown imaging algorithm, and the radar image as far as possible close to real image is formed in the case of the mating display screen of shortage.
A kind of radar image simulation method based on image recognition, it is described containing main control unit and signal receiving unit
Main control unit is an industrial personal computer, main control unit built-in signal receiving unit and simulation software, and signal receiving unit is a thunder
Up to signal receiving board card, which can convert digital information simultaneously for the radar signal received under the control of simulation software
The digital information is read by simulation software；It is characterized in that：The simulation software includes having image display and image to know
Other module；Picture recognition module has and the matched character matrix of radar image；The number letter that image display is passed back from radar
The angle and color information that radar image is extracted in breath, generate the true picture part in radar image, and select with character matrix
The interpolation parameter selected generates the supplemental image part in radar image, finally by two parts image superposition, to generate finally
Height emulation radar image, is shown on the display screen of main control unit.
The determination of above-mentioned interpolation parameter comprises the following steps：1) by the multiple similar of the radar display screen for needing to emulate
Standard picture is input in simulation software, and the picture recognition module of simulation software is according to the Pixel Information matrix of each standard picture
Form the training matrix and eigenmatrix of each standard picture；2) interpolation parameter of the image display of simulation software to select
The emulating image of standard picture is formed with the default value of user's input, picture recognition module is by the instruction of emulating image and standard picture
Practice matrix and carry out the eigenmatrix that operation generates emulating image, by the eigenmatrix of the eigenmatrix of emulating image and standard picture
It compares to identify image, if the two difference is excessive, modifies interpolation parameter and regenerate emulating image, again to the two
Eigenmatrix identified, until the eigenmatrix of emulating image and the eigenmatrix of standard picture are identical, interpolation at this time
Parameter can be used to generate the high emulating image of the standard picture.
If you need to emulate to different types of radarscope, weight after the standard picture being input in simulation software is removed
Newly the standard picture for the radar display screen for needing to emulate is inputted, above-mentioned work is repeated and selects the interpolation parameter of the type radar i.e.
It can.
The beneficial effects of the present invention are：This method can carry out height to the radar image of unknown images algorithm and go back orthoform
Very, currently, the radar image of display and practical radar image are emulated on industrial personal computer has certain gap, sensitive making radar
When degree tests this kind of test for needing to show radar image, industrial personal computer emulating image often with the figure that uses airborne indicator to show
As there is certain difference, there is certain gap so as to cause measuring accuracy in ground test precision and machine, the present invention can reduce emulation
The difference of image and real image, to improve the precision of ground radar test.
Attached drawing is described in further detail the application with reference to embodiments.
Detailed description of the invention
Fig. 1 is the radar image simulation functional block diagram based on image recognition
Fig. 2 is radar image imaging schematic diagram
Fig. 3 is radar image simulation method flow diagram
Explanation is numbered in figure：1 true picture part, 2 supplemental image parts
Specific embodiment
Referring to attached drawing, a kind of radar image simulation method based on image recognition receives single containing main control unit and signal
Member, the main control unit are an industrial personal computer, main control unit built-in signal receiving unit and simulation software, signal receiving unit
It is that a radar signal receives board, which can convert number for the radar signal received under the control of simulation software
Word information simultaneously reads the digital information by simulation software；It is characterized in that：The simulation software includes having image display
And picture recognition module；Picture recognition module has and the matched character matrix of radar image；Image display is passed back from radar
Digital information in extract the angle and color information of radar image, 1 part of true picture in radar image is generated, and with number
The interpolation parameter of word matrix selection generates 2 part of supplemental image in radar image, finally by true picture 1 and supplemental image 2
Two parts image superposition, so that final high emulation radar image is generated, as shown in Fig. 2, being shown in the display screen of main control unit
On.
In embodiment, user is using microwave signal source in suitable distance with the radar working frequency range microwave irradiation of proper strength
Radar makes the radar image shown on airborne indicator have following characteristics：Radar image positioned at picture centre possesses radar
All colors of image, while radar sector image two sides various colors is reduced gradually, black is gradually increased, on sector image side
Edge black captures all images.
User uses camera or other tools, shoots the radar image of the known navascope more than 350, and
The image section other than display screen is clipped, processing image makes its resolution ratio not less than 400x300.It is set forth below for convenience, I
Such image is referred to as standard picture.
Multiple similar standard pictures of the radar display screen emulated will be needed to be input in simulation software, simulation software
Picture recognition module forms the training matrix and feature square of each standard picture according to the Pixel Information matrix of each standard picture
Battle array.
Standard picture is input to simulation software.Simulation software judges the image of input, if it does not meet the requirements,
Then not typing image and user is prompted to re-type image, until being more than 350 with the satisfactory standard picture quantity of typing.
There is character matrix inside the picture recognition module of simulation software.After typing, picture recognition module is with these standards
Parameter adjustment is carried out to internal character matrix on the basis of image.The exclusive character matrix of each standard picture is formed, for side
Just it illustrates and understands, picture recognition module is referred to as training matrix according to the character matrix that each standard picture is formed.
The parameter tuning process of training matrix is as follows：
1. picture recognition module sets the character matrix of a 400x300 identical as standard picture resolution sizes, will scheme
As upper each color individually corresponds to a number, after the pixel color on image is completely converted into number, opsition dependent one
One corresponding filling character matrix.4*4 sampling is carried out to the character matrix, so that it becomes the character matrix of a 100*75.I
Just obtained the character matrix for possessing the image all information.It is set forth below for convenience, we are by these by image
The character matrix being transformed is referred to as Pixel Information matrix.
2. after the standard picture that user inputs is completely converted into Pixel Information matrix by picture recognition module, using these pictures
Prime information matrix carries out tune ginseng to the training matrix inside picture recognition module.
Pixel Information matrix and training matrix are subjected to convolution, and it is handled to convolution results progress pond and excitation function
Full connection processing is carried out to obtained result again afterwards, so that it may obtain the processing result of a character matrix, for convenience hereafter
It illustrates, the processing result of Pixel Information matrix and training matrix is referred to as matrix of consequence by us.
After the standard picture that user inputs is completely converted into Pixel Information matrix by picture recognition module, these pixels are used
Information matrix carries out tune ginseng to the training matrix inside picture recognition module, and Pixel Information matrix is rolled up with training matrix
Product, obtains the matrix of consequence of multiple standard pictures, picture recognition module compares the matrix of consequence between various criterion image, makes it
The mutually unison of digital quantity should be greater than 90%, at this time using any one matrix of consequence as the eigenmatrix of standard picture.
It is handled, is obtained using the Pixel Information matrix and training matrix that are transformed by one of the standard picture inputted
Matrix of consequence A.Training matrix is reused after replacement Pixel Information matrix to be handled, and matrix of consequence B is obtained.Image recognition
Module comparison result matrix A and matrix of consequence B are then adjusted if number identical between matrix of consequence is less than the 90% of total quantity
The internal digital size of these training matrix.After adjustment, again with the Pixel Information matrix generated by two various criterion images
Matrix of consequence A and matrix of consequence B is generated to be compared by picture recognition module.Persistently work, until in picture recognition module
Training matrix can generate same numbers to Pixel Information matrix that the standard picture of all inputs is converted to and be greater than matrix of consequence
The matrix of consequence of numerical sum amount 90%.It is suitable to be considered as training matrix parameter for simulation software at this time, and preservation is extracted by training matrix
A matrix of consequence out is as eigenmatrix, into further work.
The principle that airborne indicator supplements radar image blank area is the color letter of record blank area two sides
Breath, then be inserted into the colouring information of two sides with special parameter between blank area, for convenience elaboration, we are by the specific ginseng
Number is known as interpolation parameter.
The image display of simulation software forms the emulation of standard picture with the default value that interpolation parameter and user input
The training matrix of emulating image and standard picture is carried out the feature square that operation generates emulating image by image, picture recognition module
Battle array, the eigenmatrix of the eigenmatrix of emulating image and standard picture is compared to identify image, if the two difference mistake
Greatly, then it modifies interpolation parameter and regenerates emulating image, the eigenmatrix of the two is identified again, until emulating image
Eigenmatrix is identical with the eigenmatrix of standard picture, and interpolation parameter at this time can be used to generate the high emulation of the standard picture
Image.
Image display first describes main straight with specified default data, then to the blank between linear interval into
Row supplement generates a sector image, which is characterized in：Image positioned at 0 degree of part of picture centre possesses radar image institute
Some colors, while sector image two sides various colors reduces black gradually and is gradually increased, at sector image edge, black is captured
All images.
.The emulating image of standard picture is namely generated with the default value of interpolation parameter.
After image display generates emulating image with the interpolation parameter value defaulted, picture recognition module is translated into picture
Prime information matrix simultaneously operates on it the matrix of consequence for obtaining simulation result with training matrix, compares the matrix of consequence and feature
Matrix, if between matrix it is identical number be less than total quantity 90%, then replace interpolation parameter, regenerate emulating image and with
The matrix of consequence and eigenmatrix that the emulating image and training matrix generate are compared, until emulating image matrix of consequence and
Identical number is greater than the 90% of total quantity between eigenmatrix.The value of interpolation parameter is saved at this time, and as ultimately generating
The interpolation parameter of emulating image.
When we obtain the value of the interpolation parameter, it is meant that high-precision image simulation can be carried out.It will be to be checked
The radar of survey is connected on main control unit by signal receiving unit, power on and open simulation software order its start emulate radar
After image, just command signal receiving unit just converts digital information for the radar signal received at regular intervals for simulation software
And the digital information is read by test software.
After reading, image display is extracted about the angle information of image and colouring information in the digital information, with this
Information generates straight line and shows on a display screen, when image display generates lower straight line, is just joined with interpolation
Several colouring informations to the blank area insertion two sides between two straight lines supplement.
This method can carry out high reduction emulation to the radar video indicator of unknown images algorithm, currently, in industrial personal computer
The radar image and practical radar image of upper emulation display have certain gap, and doing, this kind of needs of radar sensitivity test are aobvious
When showing the test of radar image, industrial personal computer emulating image often has certain difference with the image for using airborne indicator to show, from
And measuring accuracy is caused on ground test precision and machine to have certain gap, the present invention can reduce emulating image and real image
Difference, to improve the precision of ground radar test.
If you need to be emulated to different types of radar and display, after the image that removing is input in simulation software again
The screen picture input of the mating display of radar to be simulated will be needed, repeats above-mentioned work.
Claims (5)
1. a kind of radar image simulation method based on image recognition, containing main control unit and signal receiving unit, the master
Controlling unit is an industrial personal computer, main control unit built-in signal receiving unit and simulation software, and signal receiving unit is a radar
Signal receiving board card, the board can under the control of simulation software, by the radar signal received be converted into digital information and by
Simulation software reads the digital information；It is characterized in that：The simulation software includes having image display and image recognition
Module；Picture recognition module has and the matched character matrix of radar image；The digital information that image display is passed back from radar
The middle angle and color information for extracting radar image, generates the true picture part in radar image, and select with character matrix
Interpolation parameter generate radar image in supplemental image part, finally by two parts image superposition, to generate final height
Radar image is emulated, is shown on the display screen of main control unit.
2. as described in claim 1 based on the radar image simulation method of image recognition, which is characterized in that the interpolation ginseng
Several determinations comprise the following steps：1) that the multiple similar standard pictures of radar display screen emulated will be needed to be input to emulation is soft
In part, the picture recognition module of simulation software forms the instruction of each standard picture according to the Pixel Information matrix of each standard picture
Practice matrix and eigenmatrix；2) image display of simulation software forms standard with the default value that interpolation parameter and user input
The training matrix of emulating image and standard picture is carried out operation and generates emulating image by the emulating image of image, picture recognition module
Eigenmatrix, the eigenmatrix of the eigenmatrix of emulating image and standard picture is compared to identify image, if two
Person's difference is excessive, then modifies interpolation parameter and regenerate emulating image, identify again to the eigenmatrix of the two, Zhi Daofang
The eigenmatrix of true image and the eigenmatrix of standard picture are identical, and interpolation parameter at this time can be used to generate the standard picture
High emulating image.
3. as claimed in claim 1 or 2 based on the radar image simulation method of image recognition, it is characterised in that：It will need imitative
At least 350 similar standard pictures of genuine radar display screen are input in simulation software, the resolution ratio of each standard picture
Not less than 400x300.
4. as claimed in claim 1 or 2 based on the radar image simulation method of image recognition, it is characterised in that：Image recognition
Module sets a character matrix identical with standard picture resolution sizes, and each color on image is individually corresponded to one
Number, after the pixel color on image is completely converted into number, opsition dependent, which corresponds, inserts the character matrix, to the number
Word matrix carries out 4*4 sampling, so that it becomes one possesses the Pixel Information character matrix of standard picture all information 100*75.
5. as claimed in claim 1 or 2 based on the radar image simulation method of image recognition, which is characterized in that image recognition
After the standard picture that user inputs is completely converted into Pixel Information matrix by module, image is known using these Pixel Information matrixes
The training matrix of other inside modules carries out tune ginseng, and Pixel Information matrix and training matrix are carried out convolution, obtain multiple standard drawings
The matrix of consequence of picture, picture recognition module compare the matrix of consequence between various criterion image, answer the mutually unison of its digital quantity
90% should be greater than, at this time using any one matrix of consequence as the eigenmatrix of standard picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810651567.3A CN108896972A (en) | 2018-06-22 | 2018-06-22 | A kind of radar image simulation method based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810651567.3A CN108896972A (en) | 2018-06-22 | 2018-06-22 | A kind of radar image simulation method based on image recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108896972A true CN108896972A (en) | 2018-11-27 |
Family
ID=64345651
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810651567.3A Pending CN108896972A (en) | 2018-06-22 | 2018-06-22 | A kind of radar image simulation method based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108896972A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113671851A (en) * | 2021-08-13 | 2021-11-19 | 南京航空航天大学 | Adaptive unmanned vehicle simulation method based on Datalog rule |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20040092403A (en) * | 2003-04-24 | 2004-11-03 | 오재호 | System and method for processing weather data in a realtime |
| CN1702691A (en) * | 2005-07-11 | 2005-11-30 | 北京中星微电子有限公司 | Voice-based colored human face synthesizing method and system, coloring method and apparatus |
| US20060158545A1 (en) * | 2005-01-19 | 2006-07-20 | Makoto Hirai | Image display method, imaging method, and image synthesis method |
| US20090185746A1 (en) * | 2008-01-22 | 2009-07-23 | The University Of Western Australia | Image recognition |
| US20100046791A1 (en) * | 2008-08-08 | 2010-02-25 | Snap-On Incorporated | Image-based inventory control system using advanced image recognition |
| US20110172514A1 (en) * | 2008-09-29 | 2011-07-14 | Koninklijke Philips Electronics N.V. | Method for increasing the robustness of computer-aided diagnosis to image processing uncertainties |
| CN102542870A (en) * | 2010-12-22 | 2012-07-04 | 中国人民解放军空军军训器材研究所 | Radar image simulation method and device |
| CN102915432A (en) * | 2012-09-11 | 2013-02-06 | 北京铁道工程机电技术研究所有限公司 | Method and device for extracting vehicle-bone microcomputer image video data |
| CN103616671A (en) * | 2013-11-19 | 2014-03-05 | 北京航空航天大学 | Phased array radar digital simulation system and simulation method thereof |
| CN105760879A (en) * | 2016-01-14 | 2016-07-13 | 西安电子科技大学 | Fourier-Mellin transform-based image geometric matching method |
| CN106199551A (en) * | 2016-07-26 | 2016-12-07 | 深圳中钧空间科学技术有限公司 | Face battle array sweeps low altitude radar signal processing system mutually |
| GB201701788D0 (en) * | 2016-02-05 | 2017-03-22 | Bae Systems Plc | Method and apparatus for generating an image |
| CN106597398A (en) * | 2016-12-21 | 2017-04-26 | 中国航空工业集团公司雷华电子技术研究所 | Digital radar simulation system |
| CN106772293A (en) * | 2016-12-28 | 2017-05-31 | 中国航空工业集团公司西安飞机设计研究所 | A kind of Simulator of Airborne Radar |
| CN107169435A (en) * | 2017-05-10 | 2017-09-15 | 天津大学 | A kind of convolutional neural networks human action sorting technique based on radar simulation image |
| CN107462940A (en) * | 2017-07-04 | 2017-12-12 | 西安飞机工业（集团）有限责任公司 | A kind of Auto-Test System of weather radar |
| CN107688182A (en) * | 2017-07-04 | 2018-02-13 | 西安飞机工业（集团）有限责任公司 | A kind of emulation display control program for weather radar detection |
-
2018
- 2018-06-22 CN CN201810651567.3A patent/CN108896972A/en active Pending
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20040092403A (en) * | 2003-04-24 | 2004-11-03 | 오재호 | System and method for processing weather data in a realtime |
| US20060158545A1 (en) * | 2005-01-19 | 2006-07-20 | Makoto Hirai | Image display method, imaging method, and image synthesis method |
| CN1702691A (en) * | 2005-07-11 | 2005-11-30 | 北京中星微电子有限公司 | Voice-based colored human face synthesizing method and system, coloring method and apparatus |
| US20090185746A1 (en) * | 2008-01-22 | 2009-07-23 | The University Of Western Australia | Image recognition |
| US20100046791A1 (en) * | 2008-08-08 | 2010-02-25 | Snap-On Incorporated | Image-based inventory control system using advanced image recognition |
| US20110172514A1 (en) * | 2008-09-29 | 2011-07-14 | Koninklijke Philips Electronics N.V. | Method for increasing the robustness of computer-aided diagnosis to image processing uncertainties |
| CN102542870A (en) * | 2010-12-22 | 2012-07-04 | 中国人民解放军空军军训器材研究所 | Radar image simulation method and device |
| CN102915432A (en) * | 2012-09-11 | 2013-02-06 | 北京铁道工程机电技术研究所有限公司 | Method and device for extracting vehicle-bone microcomputer image video data |
| CN103616671A (en) * | 2013-11-19 | 2014-03-05 | 北京航空航天大学 | Phased array radar digital simulation system and simulation method thereof |
| CN105760879A (en) * | 2016-01-14 | 2016-07-13 | 西安电子科技大学 | Fourier-Mellin transform-based image geometric matching method |
| GB201701788D0 (en) * | 2016-02-05 | 2017-03-22 | Bae Systems Plc | Method and apparatus for generating an image |
| CN106199551A (en) * | 2016-07-26 | 2016-12-07 | 深圳中钧空间科学技术有限公司 | Face battle array sweeps low altitude radar signal processing system mutually |
| CN106597398A (en) * | 2016-12-21 | 2017-04-26 | 中国航空工业集团公司雷华电子技术研究所 | Digital radar simulation system |
| CN106772293A (en) * | 2016-12-28 | 2017-05-31 | 中国航空工业集团公司西安飞机设计研究所 | A kind of Simulator of Airborne Radar |
| CN107169435A (en) * | 2017-05-10 | 2017-09-15 | 天津大学 | A kind of convolutional neural networks human action sorting technique based on radar simulation image |
| CN107462940A (en) * | 2017-07-04 | 2017-12-12 | 西安飞机工业（集团）有限责任公司 | A kind of Auto-Test System of weather radar |
| CN107688182A (en) * | 2017-07-04 | 2018-02-13 | 西安飞机工业（集团）有限责任公司 | A kind of emulation display control program for weather radar detection |
Non-Patent Citations (1)
| Title |
|---|
| 方伟等: "基于数字地形的雷达图像仿真系统设计", 《现代雷达》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113671851A (en) * | 2021-08-13 | 2021-11-19 | 南京航空航天大学 | Adaptive unmanned vehicle simulation method based on Datalog rule |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100389437C (en) | Method and apparatus for modeling film grain patterns in the frequency domain | |
| US5054097A (en) | Methods and apparatus for alignment of images | |
| CN107172418B (en) | A kind of tone scale map image quality evaluating method based on exposure status analysis | |
| CN108428227A (en) | Non-reference picture quality appraisement method based on full convolutional neural networks | |
| CN107635136B (en) | View-based access control model perception and binocular competition are without reference stereo image quality evaluation method | |
| CN112184554A (en) | Remote sensing image fusion method based on residual mixed expansion convolution | |
| CN109801232A (en) | A kind of single image to the fog method based on deep learning | |
| CN109636927B (en) | System and method for training and identifying aircraft attitude measurement algorithm | |
| CN111383285A (en) | Millimeter wave radar and camera sensor fusion calibration method and system | |
| CN109643446A (en) | Circuit device, electronic equipment and error-detecting method | |
| CN106686281B (en) | Fuse circuit board noise suppression ability test system | |
| CN108304243A (en) | Interface creating method, device, computer equipment and storage medium | |
| CN110335330A (en) | Image simulation generation method and its system, deep learning algorithm training method and electronic equipment | |
| Lavoué et al. | Perceptual quality of BRDF approximations: dataset and metrics | |
| CN111080540B (en) | Training method of image restoration model and computer equipment | |
| CN108896972A (en) | A kind of radar image simulation method based on image recognition | |
| CN112381751A (en) | Online intelligent detection system and method based on image processing algorithm | |
| JP7463186B2 (en) | Information processing device, information processing method, and program | |
| CN110411439B (en) | Method, device and medium for generating simulation star points according to star energy levels | |
| CN110135413B (en) | Method for generating character recognition image, electronic equipment and readable storage medium | |
| CN110796635A (en) | Shear wave transformation-based light field image quality evaluation method | |
| CN116597246A (en) | Model training method, target detection method, electronic device and storage medium | |
| CN108519215B (en) | Pupil distance adaptability test system and method and test host | |
| CN110310270A (en) | Tumour microballoon profile testing method and device based on U-net network model | |
| Bosdogianni et al. | Mixed pixel classification with robust statistics |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181127 |
|
| WD01 | Invention patent application deemed withdrawn after publication |