CN108445466A - A kind of radar adjacent target dividing method based on rainy erosion algorithm - Google Patents
A kind of radar adjacent target dividing method based on rainy erosion algorithm Download PDFInfo
- Publication number
- CN108445466A CN108445466A CN201810420398.2A CN201810420398A CN108445466A CN 108445466 A CN108445466 A CN 108445466A CN 201810420398 A CN201810420398 A CN 201810420398A CN 108445466 A CN108445466 A CN 108445466A
- Authority
- CN
- China
- Prior art keywords
- radar
- adjacent target
- target
- rainy
- dividing method
- 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
- 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
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
Abstract
The invention belongs to radar data processing technology fields, disclose a kind of radar adjacent target dividing method based on rainy erosion algorithm, radar video data will be inputted after Radar Signal Processing；Requirement according to radar system to aimed at precision, by the way that in plane occupied area, dimensions length determines region shared by the adjacent target for needing to divide；The measured value of the same range cell different direction unit in the region occupied to adjacent target takes out individually processing, i.e., each range cell channel individually puts forward individually to handle, and after the completion of processing, carries out object detection process to the region again.Present invention simulation is rainy to wash away mountain range, constantly corrodes massif, forms river valley, and mountain range can be by the separated process in a rule streams, for handling the adjacent target divided radar detection and arrived by continuous corrosion, one by one mountain peak.The present invention proves in experiment and practice, works well, and meets requirement of the high-resolution radar to Target Segmentation, divides for radar adjacent target.
Description
Technical field
The invention belongs to radar data processing technology field more particularly to a kind of radar based on rainy erosion algorithm are adjacent
Target Segmentation method is split extension target for radar data processing system.
Background technology
Currently, the prior art commonly used in the trade is such：Along with the high speed development of modern radar, radar observation is differentiated
There is radar extension target detection tracking problem in the continuous improvement of rate.In high-accuracy high-resolution radar, radar video number
Region of each pixel in other words under the rectangular coordinate system represented by cell will become smaller in, so, some mesh
The size of mark (such as naval vessel) will be greater than a radar pixel cell, occupy multiple cells.Same target exist simultaneously in
In multiple radar video data pixels, the area block with definite shape feature is formd.When being measured for intensive adjacent target,
Since radar beam has one fixed width and Radar Target Scatter, in the intermediate region of multiple mutual adjacent targets, at this
Measured value, while being influenced by adjacent radar target, cause measured value to become larger so that can not be split between target.
If not doing specially treated to measured value in this case, two or more target can be considered as a target output.It is defeated
This measured value for going out is simultaneously inaccurate.This phenomenon is particularly evident in radar near region, heavy dense targets region (navigation channel or harbour).
The measured value of multiple targets, which can mix, is taken as a target or a clutter area to handle, this will result in system compared with
Big error.Generally adjacent target was not divided in the past, is considered as a target output, or use for reference the multiple target in image procossing
Dividing method, such as intensity slicing algorithm, texture segmentation algorithm and watershed algorithm etc., but radar video data are with image
Data have difference, some are simultaneously not suitable for, or deal with ineffective.Radar system in the case of raising is intensive to adjacent target
Targeting accuracy is detected, for example to navigation channel, the becoming for adjacent target segmentation detection under ship docks, fleet compels to be essential
It solves the problems, such as.
In conclusion problem of the existing technology is：Measured value becomes larger when intensive adjacent target measures so that target it
Between can not divide, measured value is simultaneously inaccurate.
Solve the difficulty and meaning of above-mentioned technical problem：Intersect navigation in tracking radar target or form into columns and navigates by water, mutually
When tracking is driven, since distance is closer between single goal close to each other, if do not divided correctly to target, it will cause
Targets Dots are interrupted, and close single goal flight path is difficult to remain stable.How to maintain to carry out tenacious tracking to these single goals, protect
Tracking accuracy is demonstrate,proved, is always that target following filtering field needs the technical barrier solved.
To multiple targets, the target when target is intersected accurately is divided, and is to ensure that obtaining single goal stablizes flight path
Premise.Single goal is carried out after accurately tracking, single goal is identified, Threat sequence, followed by Accurate Interception
Strike, plays the role of the commanding and decision-making of commander in information war very important.
Invention content
In view of the problems of the existing technology, the present invention provides a kind of radar adjacent targets based on rainy erosion algorithm
Dividing method.
The invention is realized in this way a kind of radar adjacent target dividing method based on rainy erosion algorithm, the base
It radar video data will be inputted after Radar Signal Processing in the radar adjacent target dividing method for the erosion algorithm that rains；According to
Requirement of the radar system to aimed at precision, by the way that in plane occupied area, dimensions length determines the adjacent target institute for needing to divide
Occupied area domain；The measured value of the same range cell different direction unit in the region occupied to adjacent target takes out individually processing, i.e.,
Each range cell channel individually puts forward individually to handle.
Assuming that thering is unknown number purpose to extend target in monitoring region, it is assumed that there is m target to be present in kth frame, by these mesh
Target state is defined as：Stm k=(xm k,vxm k,ym k,vym k,lm,wm), 1≤m≤M. wherein xm k,ym k, m-th of target of expression
Position of the barycenter in kth frame under x-y rectangular coordinate systems, vxm k,vym k, indicate the movement in m-th of target target kth frame
Speed.lm,wm, indicate the long axis and short axle of the target, it is considered that target size does not change over time, for ellipse knot
Structure.Required precision determines that the long axis of single target and short axle maximum value are respectively l in radar systemmax、wmax, after target detection
It was found that the target size long axis l detectedm>lmaxOr short axle wm>lmaxOr occupied area is more than single target maximum value, then
Can be determined that this not is a single goal, and target detection is carried out to the region.
Further, the radar adjacent target dividing method based on rainy erosion algorithm includes the following steps：
Step 1 inputs radar video data, radar video data will be inputted after Radar Signal Processing；Radar video
There are Na × Nr cells for each frame of data, and Na and Nr are that azimuth axis and distance axis cell are total respectively, Zp=z (r,
A), 1≤r≤Nr, 1≤a≤Na } indicate the measured values of each coordinate unit lattice, r, a indicate that distance axis and azimuth axis are sat respectively
Mark, by distance axis, azimuth axis is regarded as a plane, which is considered as ground Plain, and there are m radar targets in cell, then
M massif can be formed on ground, the amplitude measurements of cell shared by radar target are considered as the height above sea level of massif；
Step 2 determines the adjacent target region of segmentation, the requirement according to radar system to aimed at precision, by plane
Occupied area, dimensions length determine region shared by the adjacent target for needing to divide；
Step 3, corrosion segmentation radar adjacent target of raining, the same range cell in the region occupied to adjacent target is not
Measured value with localizer unit takes out individually processing, each range cell channel individually puts forward individually to handle.
Further, the step 3 specifically includes：
(1) estimate that raindrop quantity, raindrop quantity are proportional to the sum of measured value range value in range cell channel；
(2) corrode, continuous corrosive attack mountain range of raining；
(3) it detects, two targets detected respectively.
Further, described (2) specifically include：The corrosion strength for defining raindrop reduces the echo amplitude of cell herein
Amount；The height above sea level that the corrosion strength of raindrop is proportional between raindrop starting and end position is poor；Next it repeats to rain-corrode, until
All raindrop have all been fallen, and the saddle between two peaks can be formed the separated mountain valley in two mountain peaks.
Another object of the present invention is to provide divided based on erosion algorithm radar adjacent target of raining described in a kind of application
The radar data processing system of method.
In conclusion advantages of the present invention and good effect are：
The imitative corrosion mountain range of raining of the present invention is used for handling due to the problem of can not dividing close to each other between target.Target is accurate
True segmentation, determine radar data processing system whether can correctly carry out targetpath starting and termination, data interconnection, with
Track.
Target Segmentation is the important component correctly handled radar data, adjacent in multiple targets and multiple target
When, effective segmentation can accurately just carry out each single goal correctly point boat association, determine the flight path of single goal.It could be right
Single goal progress accurately tracks, target is identified, Threat sorts, and carries out precision strike to target, is correctly determined in battlefield
Plan commander and indispensable a key technology of institute of gaining mastery over the enemy.
The present invention proves in experiment and practice, works well, and meets requirement of the high-resolution radar to Target Segmentation,
Divide for radar adjacent target.
Description of the drawings
Fig. 1 is provided in an embodiment of the present invention based on the erosion algorithm radar adjacent target dividing method flow chart that rains.
Fig. 2 is provided in an embodiment of the present invention based on erosion algorithm radar adjacent target dividing method implementation process of raining
Figure.
Fig. 3 is the inswept two mutual adjacent targets of radar beam provided in an embodiment of the present invention.
Fig. 4 is the measured value that the mutual adjacent target of inswept two of radar beam provided in an embodiment of the present invention generates.
Fig. 5 is rainy erosion algorithm corrosion flow diagram provided in an embodiment of the present invention.
Fig. 6 is radar real video datagram provided in an embodiment of the present invention.
Fig. 7 is result figure after the rainy erosion algorithm processing of radar real video data provided in an embodiment of the present invention.
Specific implementation mode
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to embodiments, to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
When the present invention can be used for Radar Targets'Detection tracking, to adjacent target by corrosion of raining radar target point
It cuts out.Improve precision of the radar system to target detection and tracking.
As shown in Figure 1, provided in an embodiment of the present invention include based on the erosion algorithm radar adjacent target dividing method that rains
Following steps：
S101：It radar video data will be inputted after Radar Signal Processing；
S102：Requirement according to radar system to aimed at precision is needed by the way that in plane occupied area, dimensions length determines
Region shared by the adjacent target of segmentation；
S103：The measured value of the same range cell different direction unit in the region occupied to adjacent target takes out individually place
Reason, i.e., each range cell channel individually puts forward individually to handle.
The application principle of the present invention is further described below in conjunction with the accompanying drawings.
Embodiment 1
In recent years, domestic and foreign scholars were proposed various imitative using simulation nature biotechnology population habit or natural phenomena
It is raw to learn algorithm.In order to divide adjacent target measured value, present invention simulation, which is rained, washes away mountain range, forms the process in river valley, it is proposed that
Rain erosion algorithm.
Mountain peak and mountain valley are very common in nature.Some mountain valleys are due to made of the river degradation between mountain peak.
And the water source of these small rivers is typically the raindrop fallen on these mountain peaks.Raindrop are to fall on massif at random, due to gravity
Effect with the shape of massif is flowed down along massif.A large amount of raindrop are flowed down along massif gradually pools streamlet, and streamlet is again
It constantly converges in larger river.The terminal in river is typically the Plain outside barrier lake or mountain range.Mountain peak was washed away in streams
Afterwards, some soil can be taken away or carry out certain corrosion to massif, generate souring so that the streams between mountain peak becomes
It is deeper, while saddle part may also form new mountain valley.Finally with raindrop erosion with wash away, mountain peak can be by a rule streams
It is separated.
Fig. 2 is particular flow sheet of the present invention；Radar plot condensing method the present invention is based on Contour extraction includes as follows
Step：Radar video data are inputted, determine region shared by the adjacent target of segmentation, adjacent radar target is divided in corrosion of raining.
(1) radar video data are inputted
It radar video data will be inputted after Radar Signal Processing；There are Na × Nr for each frame of radar video data
Cell, Na and Nr are that azimuth-range cell is total respectively, and Zp={ z (r, a), 1≤r≤Nr, 1≤a≤Na } indicate every
The measured value of a coordinate unit lattice, r, a indicate distance axis and orientation axial coordinate respectively, and here by distance axis, azimuth axis is regarded as one
The amplitude measurements of cell are considered as the height above sea level of " massif " by a plane, if there are m targets in cell, in three-dimensional
There are m mountain peaks in plane.
(2) the adjacent target region of segmentation is determined
When distance is distant between radar target, one by one radar target formed in distance, orientation and measured value three
Just as independent mountain peak one by one in dimensional plane, as shown in figure 3, due to radar beam has one fixed width, such as Fig. 4 institutes
Show, the measured value of same radar video cell is possible to be influenced by two adjacent targets close to each other simultaneously, distance
Closer adjacent target forms mountain range just as rolling mountain peak.The plane that mountain range is formed in distance axis and azimuth axis occupies
Size be much larger than the area that occupies of mountain peak, while existing also greater than single mountain peak in distance axis and the length scale of azimuth axis
Length on the two axis.Requirement according to system to aimed at precision, area, dimensions length determine the adjacent target area of segmentation
Domain.
(3) rain the adjacent radar target of segmentation.
Due to the effect of radar beam, measured value correlation in orientation is larger.Therefore same distance unit different direction list
The measured value of member takes out individually processing, each range cell channel proposes individually processing.
Processing step is divided into three steps.The first step is estimation raindrop quantity, and raindrop quantity is proportional in range cell channel and surveys
The sum of magnitude amplitude；
Next by one drop raindrop for illustratively rain process.N is arrived in the initial position of raindrop 1aIt is upper uniformly random
Distribution.Assuming that starting point is a, then (the initial position of a, z (r, a)) expression raindrop is used.After finding initial position, raindrop by
Lower position is constantly looked for gravity, finds the lower unit of echo amplitude near current location always in other words
Lattice, until raindrop flow into " Plain " or " barrier lake ".There are two types of the final results of raindrop, if raindrop reach " Plain ", is regarded as
It disappears, does not accumulate.If raindrop reach " barrier lake ", the horizontal plane of " barrier lake " can rise.Once the water of " barrier lake " is full
, in other words the horizontal plane of " barrier lake " be higher than surrounding, new " raindrop " can flow away.The process of algorithm is exactly to simulate nature
The rainy process in boundary；
Second step, corrosion, the lower place between water constantly flows through mountain can be washed away and generate deeper and deeper mountain valley, this
It is exactly corrosion of the water to massif.In the present invention, the power of corrosiveness is decided by the raindrop on the cell, corrodes to cell
Functioning as reduces the echo amplitude of cell herein.The corrosion strength for defining raindrop, that is, reduce the echo width of cell herein
The amount of degree.The height above sea level that the corrosion strength of raindrop is proportional between raindrop starting and end position is poor, in other words raindrop start element lattice
And the difference of terminal cell echo amplitude.Next this process of raining-corrode is repeated, until all " raindrop " have all been fallen
Under.Along with this process, the saddle between two peaks can be formed can be by the separated mountain valley in two mountain peaks.
The visible Fig. 5 of this complete procedure.In Figure 5, data source is a part of measured data taken out, two targets
Measured value mixes, it is difficult to differentiate.It is like two mountain peaks from image.During algorithm iteration, produce
Two barrier lakes, wherein barrier lake 1 are on mountain top, and barrier lake 2 is in saddle.Flow into barrier lake 1 raindrop drop it is small, corrosion strength compared with
Small, there is no generate larger change to echo amplitude herein.And barrier lake 2 is then different, 2 moderate rain of barrier lake drippage difference is big, rotten
It is strong to lose power, constantly corrodes saddle, produces a mountain valley, it can be separated by two mountain peaks.Several times by this after iteration
Measured value on range cell lattice is put back in video data, the video data of the result such as lower rights Fig. 5 displaying.It can be apparent
See, the measured value amplitude between two targets becomes smaller.In detection process, they will be detected as two individually
Target rather than a larger target.
Third walks, detection, and along with this process is corroded, the saddle between two peaks can be formed can be by two mountain peaks point
The mountain valley isolated.The measured value amplitude being equivalent between this adjacent target is lower, and forms independent mountain peak, is detecting
When, so that it may radar target detected respectively.For subsequent radar plot agglomeration process and Track In Track.
Embodiment 2
Based on the radar adjacent target dividing method of rainy erosion algorithm with embodiment 1
It tests below and the effect of the present invention is further described
Measured data
Carry out the effect of illustratively the method for the present invention using measured data.Fig. 6 is the part extracted in measured data.
Abscissa indicates Azimuth ＆ Range axis respectively with ordinate.Color indicates that measured value increases from black to white in Fig. 6 and Fig. 7, and Fig. 6 is
Extract untreated part measured value, Wei Baichushi clutters area；Fig. 7 is by clutter processing, target detection and corruption of raining excessively
The measured value retained after erosion algorithm；Finally, a total of 8 targets, are marked by jack per station.Wherein 3 He of target
Target 4 and target 7 and target 8 are in same range cell, and orientation is at a distance of relatively close, some measured value between two targets
Kept amplitude higher by two object effects simultaneously, the profile of target 3 and target 4 is connected together, target 7 and target 8
Profile, which connects together, to be estimated, to be a larger target by two target detections, to influence measurement accuracy.From Fig. 7
As can be seen that after being corrected by the measured value for the erosion algorithm that rains, between target 3 and target 4, between target 7 and target 8
The higher measured value of amplitude is removed.So, no matter which kind of contour following algorithm is used, it all can be by the two targets
Profile detected respectively, such as Fig. 7, detected all single goals to detect.Here it is the present invention to mutually adjacent more
The improvement that target is done.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
All any modification, equivalent and improvement etc., should all be included in the protection scope of the present invention made by within refreshing and principle.
Claims (5)
1. a kind of radar adjacent target dividing method based on rainy erosion algorithm, which is characterized in that described based on corrosion of raining
The radar adjacent target dividing method of algorithm radar video data will input after Radar Signal Processing；According to radar system pair
The requirement of aimed at precision, by the way that in plane occupied area, dimensions length determines region shared by the adjacent target for needing to divide；To phase
The measured value of the same range cell different direction unit in the region that adjacent target is occupied takes out individually processing, i.e., each distance is single
First channel individually puts forward individually to handle.
2. the radar adjacent target dividing method based on rainy erosion algorithm as described in claim 1, which is characterized in that described
Radar adjacent target dividing method based on rainy erosion algorithm includes the following steps：
Step 1 inputs radar video data, radar video data will be inputted after Radar Signal Processing；Radar video data
Each frame there are Na × Nr cell, Na and Nr are azimuth axis and distance axis cell sum respectively, Zp=z (r, a), 1
≤ r≤Nr, 1≤a≤Na } indicate that the measured values of each coordinate unit lattice, r, a indicate distance axis and orientation axial coordinate respectively, it will
Distance axis, azimuth axis are regarded as a plane, which is considered as ground Plain, and there are m radar targets in cell, then can be on ground
Face forms m massif, and the amplitude measurements of cell shared by radar target are considered as the height above sea level of massif；
Step 2 determines cut zone, the requirement according to radar system to aimed at precision, by plane occupied area, size
Length determines region shared by the adjacent target for needing to divide；
Step 3, corrosion segmentation radar adjacent target of raining, the same range cell in the region that adjacent target is occupied not Tongfang
The measured value of bit location takes out individually processing, each range cell channel individually puts forward individually to handle.
3. the radar adjacent target dividing method based on rainy erosion algorithm as claimed in claim 2, which is characterized in that described
Step 3 specifically includes：
(1) estimate that raindrop quantity, raindrop quantity are proportional to the sum of measured value range value in range cell channel；
(2) corrode, continuous corrosive attack mountain range of raining；
(3) it detects, adjacent target detected respectively.
4. the radar adjacent target dividing method based on rainy erosion algorithm as claimed in claim 3, which is characterized in that described
(2) it specifically includes：The corrosion strength for defining raindrop, that is, reduce the amount of the echo amplitude of cell herein；The corrosion strength direct ratio of raindrop
Height above sea level between raindrop starting and end position is poor；Next it repeats to rain-corrode, until all raindrop have all been fallen,
Saddle between two peaks can be formed the separated mountain valley in two mountain peaks.
5. a kind of radar adjacent target dividing method using based on rainy erosion algorithm described in Claims 1 to 4 any one
Radar data processing system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810420398.2A CN108445466B (en) | 2018-05-04 | 2018-05-04 | Radar adjacent target segmentation method based on rain erosion algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810420398.2A CN108445466B (en) | 2018-05-04 | 2018-05-04 | Radar adjacent target segmentation method based on rain erosion algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108445466A true CN108445466A (en) | 2018-08-24 |
| CN108445466B CN108445466B (en) | 2021-09-28 |
Family
ID=63203003
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810420398.2A Active CN108445466B (en) | 2018-05-04 | 2018-05-04 | Radar adjacent target segmentation method based on rain erosion algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108445466B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103996039A (en) * | 2014-05-06 | 2014-08-20 | 河海大学 | SAR image channel extraction method combining gray-level threshold-value segmentation and contour shape identification |
| CN105374033A (en) * | 2015-10-19 | 2016-03-02 | 西安电子科技大学 | SAR image segmentation method based on ridgelet deconvolution network and sparse classification |
| KR20160053520A (en) * | 2014-11-05 | 2016-05-13 | 국방과학연구소 | Efficient Method and Apparatus of Generating Inverse Synthetic Aperture Radar Image of Multiple Targets Using Flight Trajectory and Morphological Processing |
| CN106372362A (en) * | 2016-09-27 | 2017-02-01 | 铁道第三勘察设计院集团有限公司 | Remote sensing technology-based loess gully headward erosion range predicting method |
| CN106803078A (en) * | 2017-01-18 | 2017-06-06 | 哈尔滨工业大学 | A kind of SAR image Ship Target dividing method |
| CN106970375A (en) * | 2017-02-28 | 2017-07-21 | 河海大学 | A kind of method that building information is automatically extracted in airborne laser radar point cloud |
| CN107274417A (en) * | 2017-07-05 | 2017-10-20 | 电子科技大学 | A kind of single wooden dividing method based on airborne laser point cloud aggregation |
| CN107918126A (en) * | 2017-11-17 | 2018-04-17 | 西安电子科技大学 | Clutter suppression method is obscured based on the multichannel inshore that multiple features are split automatically |
-
2018
- 2018-05-04 CN CN201810420398.2A patent/CN108445466B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103996039A (en) * | 2014-05-06 | 2014-08-20 | 河海大学 | SAR image channel extraction method combining gray-level threshold-value segmentation and contour shape identification |
| KR20160053520A (en) * | 2014-11-05 | 2016-05-13 | 국방과학연구소 | Efficient Method and Apparatus of Generating Inverse Synthetic Aperture Radar Image of Multiple Targets Using Flight Trajectory and Morphological Processing |
| CN105374033A (en) * | 2015-10-19 | 2016-03-02 | 西安电子科技大学 | SAR image segmentation method based on ridgelet deconvolution network and sparse classification |
| CN106372362A (en) * | 2016-09-27 | 2017-02-01 | 铁道第三勘察设计院集团有限公司 | Remote sensing technology-based loess gully headward erosion range predicting method |
| CN106803078A (en) * | 2017-01-18 | 2017-06-06 | 哈尔滨工业大学 | A kind of SAR image Ship Target dividing method |
| CN106970375A (en) * | 2017-02-28 | 2017-07-21 | 河海大学 | A kind of method that building information is automatically extracted in airborne laser radar point cloud |
| CN107274417A (en) * | 2017-07-05 | 2017-10-20 | 电子科技大学 | A kind of single wooden dividing method based on airborne laser point cloud aggregation |
| CN107918126A (en) * | 2017-11-17 | 2018-04-17 | 西安电子科技大学 | Clutter suppression method is obscured based on the multichannel inshore that multiple features are split automatically |
Non-Patent Citations (2)
| Title |
|---|
| LU, PP ET AL.: "A New Segmentation Method for Connected Characters in CAPTCHA", 《FOURTH INTERNATIONAL CONFERENCE ON CONTROL, AUTOMATION AND INFORMATION SCIENCES》 * |
| 朱兰 等: "基于滴水算法的印刷体维吾尔文切分方法", 《计算机技术与发展》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108445466B (en) | 2021-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105182328B (en) | A kind of GPR buried target detection method based on two-dimensional empirical mode decomposition | |
| CN102999914B (en) | Automatic recognition method of continental slope foot point based on terrain grid | |
| Soomere et al. | Towards identification of areas of reduced risk in the Gulf of Finland, the Baltic Sea. | |
| Huang et al. | Risk assessment of ships maneuvering in an approaching channel based on AIS data | |
| CN102109597B (en) | Method for identifying type of ship according to high-resolution SAR (synthetic aperture radar) image of ship | |
| CN108957437B (en) | Radar target tracking method based on environmental situation analysis | |
| CN106569193B (en) | The small targets detection in sea clutter method filtered based on anterior-posterior to income reference particle | |
| CN101377545B (en) | Radar device | |
| CN104361582A (en) | Method of detecting flood disaster changes through object-level high-resolution SAR (synthetic aperture radar) images | |
| CN104331886A (en) | Port region ship and warship detection method based on high resolution SAR image | |
| CN109143182B (en) | A kind of multiple dimensioned radar block plan immediate processing method | |
| CN107862271A (en) | A kind of detection method of Ship Target | |
| CN105866758A (en) | Time frequency double feature sea surface small target detection method based on block whitening clutter suppression | |
| CN104391294A (en) | Connection domain characteristic and template matching based radar plot correlation method | |
| CN104537675A (en) | SAR image of bilateral CFAR ship target detection method | |
| CN104007428B (en) | A kind of method identifying the non-meteorological echo that running car causes in radar return | |
| CN107835998B (en) | Hierarchical tiling method for identifying surface types in digital images | |
| CN116482644B (en) | Sea fog identification method | |
| CN108445466A (en) | A kind of radar adjacent target dividing method based on rainy erosion algorithm | |
| CN105551049A (en) | Port water area extraction based large port detection method for polarimetric SAR image | |
| Westfeld et al. | Analysis of the effect ofwave patterns on refraction in airborne lidar bathymetry | |
| Puissant et al. | Coastline extraction in VHR imagery using mathematical morphology with spatial and spectral knowledge | |
| CN111105419A (en) | Vehicle and ship detection method and device based on polarized SAR image | |
| Grünthal et al. | Monitoring of coastal processes by using airborne laser scanning data | |
| CN114565176A (en) | Long-term ship track prediction 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |