CN109186533A - A kind of ground air defense radar shield angle calculation method based on BP neural network algorithm - Google Patents
A kind of ground air defense radar shield angle calculation method based on BP neural network algorithm Download PDFInfo
- Publication number
- CN109186533A CN109186533A CN201810770794.8A CN201810770794A CN109186533A CN 109186533 A CN109186533 A CN 109186533A CN 201810770794 A CN201810770794 A CN 201810770794A CN 109186533 A CN109186533 A CN 109186533A
- Authority
- CN
- China
- Prior art keywords
- radar
- neural network
- shield angle
- output
- training
- 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
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C1/00—Measuring angles
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a kind of, and the ground air defense radar shield angle calculation method based on BP neural network algorithm obtains the echo data of radar with radar beam scanning search screen；The detection of MTD moving-target is done to radar return, extract search shield in static veil information, azimuth, pitch angle including veil and the distance between with observation point constitute radar shield angle sample set；Select a radar shield angle sample as test sample, remaining composition training set is normalized respectively；It is input with the azimuth of training sample, the shield angle, distance on azimuth are output, training BP neural network model；The BP neural network model of training is tested using test sample, if the error of model output and normalized test sample actual value is less than the threshold value of setting, otherwise output model output, i.e. azimuth-masking angular curve resurvey.The present invention improves the efficiency and precision of ground air defense radar shield angle calculating, adapts to changeable terrain environment.
Description
Technical field
The present invention relates to air defense radar shield angle rendering techniques, and in particular to a kind of ground based on BP neural network algorithm
Air defense radar shield angle calculation method.
Background technique
Radar finds the minimum elevation of target shield angle in a certain direction.When the angle of site of radar antenna is less than this
When a angle, due to blocking for landform or atural object, radar can not find target.For the radar of maneuver warfare, how
Radar periphery veil situation is quickly grasped, radar shield angle figure is drawn, is an important process of air defense operational control.
Traditional shield angle measurement method uses optical measurement, troublesome in poeration although reliability is higher, the time-consuming a few days.
In recent years, 2013 " GIS-Geographic Information System is applied in gun bore detection ", sharp with the foundation of Distribution GIS
Automatic elevation amendment, shield angle computing function are completed with landform elevation information.But the data inputting of GIS-Geographic Information System, still uses
Traditional mapping mode, wartime terrain environment is changeable, and gis database is difficult to timely update, and air defense radar is to masking
Angular data accuracy requirement is stringent, and GIS is not able to satisfy the demand of motor-driven air defense radar." The Radar Screen in 2016
Angle Automatic Measurement System_Kangkang Yin " in propose radar be equipped with optical device, obtain ground
Horizontal line image joins line using image Segmentation Technology identification vacant lot.But this method there are be difficult to when soft edge, nothing
Method excludes the technical problems such as moving-target interference, and visible light operating distance is close when optical imagery, is influenced by weather.
Summary of the invention
The demand for obtaining terrain information as early as possible is needed after setting up for high maneuverability air defense radar, it is an object of the invention to mention
For a kind of ground air defense radar shield angle calculation method based on BP neural network algorithm, ground air defense radar shield angle is improved
The efficiency and precision of calculating.
The technical solution for realizing the aim of the invention is as follows: a kind of ground air defense radar screening based on BP neural network algorithm
Angle calculation method is covered, is included the following steps:
Step 1, radar beam scanning search screen, obtain the echo data of radar；
Step 2 does the detection of MTD moving-target to radar return, extracts search and shields interior static veil information, including veil
Azimuth, pitch angle and the distance between with observation point, constitute radar shield angle sample set；
Step 3 selects a radar shield angle sample as test sample, remaining composition training set carries out normalizing respectively
Change processing；
Step 4 with the azimuth of training sample after normalizing is input, and the shield angle, distance on azimuth are output, instruction
Practice BP neural network model；
Step 5, the BP neural network model that training is tested using the test sample after normalization, obtain the mould of test sample
Type output, if threshold value of the model output with the error of normalized test sample actual value less than setting, output model output,
That is otherwise azimuth-masking angular curve goes to step 1 and resurveys.
Compared with prior art, the present invention its remarkable advantage are as follows: 1) present invention, can be full-automatic by radar without manual operation
It realizes, from acquisition data to output shield angle time-consuming control in 10s, compared to traditional handwork, efficiency is greatly improved；
2) the method for the present invention booting after can real-time measurement, compared to based on modernization gis database calculation method,
It is suitable for changeable wartime terrain environment；3) present invention joins line compared to image Segmentation Technology identification vacant lot, to light and
The robustness of weather is more preferable.
Detailed description of the invention
Fig. 1 is the ground air defense radar shield angle calculation method flow chart the present invention is based on BP neural network algorithm.
Specific embodiment
As shown in Figure 1, the ground air defense radar shield angle calculation method based on BP neural network algorithm, including walk as follows
It is rapid:
1) data acquire: using the entire search screen of radar beam scanning, obtaining radar return data.Optics is replaced with radar
Measurement, operating distance is remote, without additional optical imaging apparatus.Sequential scan entirely searches for screen, and acquired results are one group of data,
To reduce error, false-alarm false dismissal is removed, acquisition order multi-group data is needed, sampling time-consuming can be according to search screen size control several
Second.
2) data prediction: MTD moving-target detection algorithm is carried out to radar return, excludes the influence of moving object, is extracted
Search shield in static veil information, azimuth, pitch angle including veil and the distance between with observation point constitute radar
Shield angle sample set.
3) sample packet: selecting a radar shield angle sample as test sample, remaining composition training set carries out respectively
Normalized, the index without dimension value being converted on (- 1,1).
4) construct and training network: the azimuth with training sample after normalizing is input, shield angle on azimuth, away from
From to export, BP neural network model is trained.As a kind of specific embodiment, 3 layers of BP neural network, input unit number are constructed
1, output unit number 2, hidden layer unit number takes 5 layers, activation primitiveWherein x is the input of neuron, f (x)
For the output of neuron.
5) verification result: testing the BP neural network model of training using the test sample after normalization, avoids that net occurs
Network falls into local minimum, acquisition data have the special circumstances such as large error or mistake.When verifying, with test sample after normalization
Azimuth be input, obtain the model output of test sample, the i.e. corresponding shield angle in the azimuth and distance；If model exports
It is less than the threshold value of setting with the error of normalized test sample actual value, then output model exports, i.e., azimuth-shield angle is bent
Otherwise line goes to step 1 and resurveys.Such as flying bird will not be used as veil after moving-target detects.If but having fortune
Move more slow object, the detection identification of non-passive target, and mistakenly by as occurring void once in a while at veil or letter
Alert, false dismissal, and cause to cover that angular data is inconsistent, changes with acquisition time, then network when may cause trained in multiple groups sample
It can not restrain, or reality output and desired output error are excessive when test verifying, need to resurvey data.
Claims (2)
1. a kind of ground air defense radar shield angle calculation method based on BP neural network algorithm, which is characterized in that including as follows
Step:
Step 1, radar beam scanning search screen, obtain the echo data of radar；
Step 2 does the detection of MTD moving-target to radar return, extracts search and shields interior static veil information, the side including veil
Parallactic angle, pitch angle and the distance between with observation point, constitute radar shield angle sample set；
Step 3 selects a radar shield angle sample as test sample, and place is normalized in remaining composition training set respectively
Reason；
Step 4 with the azimuth of training sample after normalizing is input, and the shield angle, distance on azimuth are output, training BP
Neural network model；
Step 5, the BP neural network model that training is tested using the test sample after normalization, the model for obtaining test sample are defeated
Out, if the error of model output and normalized test sample actual value is less than the threshold value of setting, output model is exported, i.e., just
Otherwise parallactic angle-masking angular curve goes to step 1 and resurveys.
2. the ground air defense radar shield angle calculation method according to claim 1 based on BP neural network algorithm, special
Sign is, step 4 constructs 3 layers of BP neural network, input unit number 1, output unit number 2, and hidden layer unit number takes 5 layers, activation
Function isWherein x is the input of neuron, and f (x) is the output of neuron.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810770794.8A CN109186533A (en) | 2018-07-13 | 2018-07-13 | A kind of ground air defense radar shield angle calculation method based on BP neural network algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810770794.8A CN109186533A (en) | 2018-07-13 | 2018-07-13 | A kind of ground air defense radar shield angle calculation method based on BP neural network algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109186533A true CN109186533A (en) | 2019-01-11 |
Family
ID=64936122
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810770794.8A Pending CN109186533A (en) | 2018-07-13 | 2018-07-13 | A kind of ground air defense radar shield angle calculation method based on BP neural network algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109186533A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111610521A (en) * | 2020-05-27 | 2020-09-01 | 西安长远电子工程有限责任公司 | Radar terrain data processing method |
| CN111722195A (en) * | 2020-06-29 | 2020-09-29 | 上海蛮酷科技有限公司 | Radar occlusion detection method and computer storage medium |
| CN114509042A (en) * | 2020-11-17 | 2022-05-17 | 易图通科技（北京）有限公司 | Shielding detection method, shielding detection method of observation route and electronic equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104021315A (en) * | 2014-06-27 | 2014-09-03 | 四川电力设计咨询有限责任公司 | Method for calculating station service power consumption rate of power station on basis of BP neutral network |
| CN104535997A (en) * | 2015-01-08 | 2015-04-22 | 西安费斯达自动化工程有限公司 | Image/laser ranging/ low-altitude pulse radar integrated system |
| CN105651267A (en) * | 2016-03-21 | 2016-06-08 | 中国人民解放军空军装备研究院雷达与电子对抗研究所 | Radar position selection method based on three dimensional laser scanner and GIS (Geographic Information System) |
| CN105931153A (en) * | 2016-04-14 | 2016-09-07 | 湘潭大学 | Indirect questionnaire assessment method based on neural network prediction analysis model |
| CN107167824A (en) * | 2017-07-26 | 2017-09-15 | 天津博创金成技术开发有限公司 | A kind of Beidou satellite navigation system quick satellite selection method |
| CN108051813A (en) * | 2017-12-04 | 2018-05-18 | 湖南华诺星空电子技术有限公司 | For the radar-probing system and method for low latitude multiple target Classification and Identification |
| CN207440280U (en) * | 2017-10-23 | 2018-06-01 | 西安长远电子工程有限责任公司 | A kind of autonomous controllable general purpose radar terminal system |
-
2018
- 2018-07-13 CN CN201810770794.8A patent/CN109186533A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104021315A (en) * | 2014-06-27 | 2014-09-03 | 四川电力设计咨询有限责任公司 | Method for calculating station service power consumption rate of power station on basis of BP neutral network |
| CN104535997A (en) * | 2015-01-08 | 2015-04-22 | 西安费斯达自动化工程有限公司 | Image/laser ranging/ low-altitude pulse radar integrated system |
| CN105651267A (en) * | 2016-03-21 | 2016-06-08 | 中国人民解放军空军装备研究院雷达与电子对抗研究所 | Radar position selection method based on three dimensional laser scanner and GIS (Geographic Information System) |
| CN105931153A (en) * | 2016-04-14 | 2016-09-07 | 湘潭大学 | Indirect questionnaire assessment method based on neural network prediction analysis model |
| CN107167824A (en) * | 2017-07-26 | 2017-09-15 | 天津博创金成技术开发有限公司 | A kind of Beidou satellite navigation system quick satellite selection method |
| CN207440280U (en) * | 2017-10-23 | 2018-06-01 | 西安长远电子工程有限责任公司 | A kind of autonomous controllable general purpose radar terminal system |
| CN108051813A (en) * | 2017-12-04 | 2018-05-18 | 湖南华诺星空电子技术有限公司 | For the radar-probing system and method for low latitude multiple target Classification and Identification |
Non-Patent Citations (1)
| Title |
|---|
| 王豪: "大区域多尺度雷达遮蔽角计算关键技术研究与实现", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111610521A (en) * | 2020-05-27 | 2020-09-01 | 西安长远电子工程有限责任公司 | Radar terrain data processing method |
| CN111722195A (en) * | 2020-06-29 | 2020-09-29 | 上海蛮酷科技有限公司 | Radar occlusion detection method and computer storage medium |
| CN111722195B (en) * | 2020-06-29 | 2021-03-16 | 江苏蛮酷科技有限公司 | Radar occlusion detection method and computer storage medium |
| CN114509042A (en) * | 2020-11-17 | 2022-05-17 | 易图通科技（北京）有限公司 | Shielding detection method, shielding detection method of observation route and electronic equipment |
| CN114509042B (en) * | 2020-11-17 | 2024-05-24 | 易图通科技（北京）有限公司 | Shading detection method, shading detection method of observation route and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108416378B (en) | Large-scene SAR target recognition method based on deep neural network | |
| Hou et al. | Deep learning-based subsurface target detection from GPR scans | |
| US8144937B2 (en) | System and method for airport mapping database automatic change detection | |
| CN110889324A (en) | Thermal infrared image target identification method based on YOLO V3 terminal-oriented guidance | |
| CN112183432B (en) | Building area extraction method and system based on medium-resolution SAR image | |
| CN108052940A (en) | SAR remote sensing images waterborne target detection methods based on deep learning | |
| CN105809194B (en) | A kind of method that SAR image is translated as optical image | |
| CN110910341B (en) | Method and device for detecting defects of rusted areas of power transmission line | |
| CN109186533A (en) | A kind of ground air defense radar shield angle calculation method based on BP neural network algorithm | |
| Shaoqing et al. | The comparative study of three methods of remote sensing image change detection | |
| CN103353988A (en) | Method for evaluating performance of heterogeneous SAR (synthetic aperture radar) image feature matching algorithm | |
| CN111553204B (en) | Transmission tower detection method based on remote sensing image | |
| CN108537169A (en) | A kind of high-resolution remote sensing image method for extracting roads based on center line and detection algorithm of having a lot of social connections | |
| CN115995005B (en) | Crop extraction method and device based on single-period high-resolution remote sensing image | |
| CN109117776A (en) | Aircraft and meteorological clutter classifying identification method based on track information | |
| CN110517228A (en) | Trunk image rapid detection method based on convolutional neural networks and transfer learning | |
| CN108073865B (en) | Aircraft trail cloud identification method based on satellite data | |
| Tang et al. | A novel sample selection method for impervious surface area mapping using JL1-3B nighttime light and Sentinel-2 imagery | |
| CN114022782B (en) | Sea fog detection method based on MODIS satellite data | |
| CN106897730A (en) | SAR target model recognition methods based on fusion classification information with locality preserving projections | |
| CN113096122A (en) | Meteor detection method and device and electronic equipment | |
| Le Bris et al. | Change detection in a topographic building database using submetric satellite images | |
| Guo et al. | A cloud boundary detection scheme combined with aslic and cnn using zy-3, gf-1/2 satellite imagery | |
| CN113627292A (en) | Remote sensing image identification method and device based on converged network | |
| CN112733661A (en) | Multi-example energy constraint minimized hyperspectral target description and detection method |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190111 |
|
| RJ01 | Rejection of invention patent application after publication |