CN107610130B - Sea-land clutter scene segmentation method based on amplitude and phase linearity ratio - Google Patents
Sea-land clutter scene segmentation method based on amplitude and phase linearity ratio Download PDFInfo
- Publication number
- CN107610130B CN107610130B CN201710720482.1A CN201710720482A CN107610130B CN 107610130 B CN107610130 B CN 107610130B CN 201710720482 A CN201710720482 A CN 201710720482A CN 107610130 B CN107610130 B CN 107610130B
- Authority
- CN
- China
- Prior art keywords
- value
- phase
- amplitude
- image
- wave position
- 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.)
- Active
Links
Images
Landscapes
- Radar Systems Or Details Thereof (AREA)
Abstract
The invention discloses a sea and land clutter scene segmentation method based on amplitude-phase linearity ratio, which comprises the steps of (1) obtaining an echo sequence, (2) establishing a distance-wave position coordinate system, (3) selecting a distance-wave position resolution unit, (4) calculating an amplitude-phase linearity matrix, (5) judging whether all points in the distance-wave position coordinate system are selected, if so, executing a step (6), otherwise, executing the step (3), (6) converting the amplitude-phase linearity matrix into a gray image, (7) generating a binary image, and (8) generating a sea and land clutter scene segmentation image.
Description
Technical Field
The invention belongs to the technical field of communication, and further relates to a sea and land clutter scene segmentation method based on amplitude-phase linearity ratio in the technical field of radar signal processing, which belongs to the technical field of .
Background
The target detection technology under the sea clutter background is crucial research directions in the radar application technology, and is generally applied in the military and civil fields, when the radar works in a sea mode, the scanning scene is complex and has a large range, radar echoes often contain various types of clutters, including sea clutters, ground clutters, reef clutter, offshore clutters and the like.
Unlike traditional image segmentation methods, the segmentation of radar clutter scenes involves converting radar data into grayscale images and segmenting the grayscale images. The doppler spectrum of the sea clutter has a larger bandwidth and the doppler spectrum of the ground clutter has a smaller bandwidth. Due to the movement of the carrier, the echo Doppler shift changes along with the azimuth angle, and the sea and land clutter scene segmentation depending on the Doppler frequency is high in calculation cost and difficult to meet the requirement of real-time sea and land clutter scene segmentation. If the clutter scene is directly depicted and segmented by using an artificial method, time and labor are wasted, huge workload is brought, and the real-time scene segmentation requirement cannot be met.
In the patent of 'FCM-based block adaptive image segmentation method' (patent application number 201310726876, publication number CN103761726B) applied by the university of river and sea, FCM-based block adaptive image segmentation methods are provided, the method respectively adopts a neighborhood average method to process an image to be segmented, then utilizes an FCM-based image segmentation method to segment the image to be segmented and a standard FCM image segmentation method to segment the image to be segmented, carries out block processing on a result image, numbers image area blocks and calculates the variance of each area block, compares the variances of the image blocks at the same position on two segmentation images, and selects an image block with a smaller variance as a final segmentation result.
In the patent ' SAR image sea-land segmentation method based on wavelet transformation and OTSU threshold' (patent application No. 201210536981.2, publication No. CN102968798B) applied by Beijing aerospace university, SAR image sea-land segmentation methods based on wavelet transformation and OTSU threshold are provided, the method utilizes the noise smoothing characteristic of wavelet transformation to suppress the speckle noise in SAR images, further adopts unsupervised and threshold-optimized OTSU threshold methods to roughly segment land areas, combines the detection results under various scales based on the multi-scale analysis characteristic of wavelet transformation, and finally obtains the final SAR image sea-land segmentation result through automatic subsequent processing and edge tracking.
Disclosure of Invention
The invention converts the echo sequence of the sea and land clutter scene received by a radar into a gray image, establishes distance-wave position coordinate systems, extracts all points in the distance-wave position coordinate systems, takes the distance-wave positions corresponding to all points as distance-wave position resolution units, and divides all the distance-wave position resolution units by using the sea and land clutter scene division method based on the ratio of the amplitude to the phase linearity, thereby achieving the purpose of sea and land clutter scene division.
The basic idea for achieving the purpose of the invention is that firstly, an echo sequence of a sea and land clutter scene received by a radar is converted into a gray level image, distance-wave position coordinate systems are established, all points in the distance-wave position coordinate systems are extracted, distance-wave positions corresponding to selected points are used as distance-wave position resolution units, then, the ratio of the amplitude and the phase linearity of each distance-wave position resolution unit is calculated to obtain an amplitude and phase linearity matrix, the matrix is converted into the gray level image, finally, a binary image after threshold segmentation is obtained by using a maximum inter-class variance method, and a 5 x 5 structural element is adopted to perform morphological filtering on the binary image to obtain a final sea and land clutter scene segmentation image.
In order to achieve the purpose, the method comprises the following specific implementation steps:
(1) acquiring an echo sequence:
receiving a pulse-distance-wave position three-dimensional echo sequence containing a sea and land clutter scene by a radar;
(2) establishing a distance-wave position coordinate system:
converting the echo sequence into a grayscale image with the size of M multiplied by L, and establishing distance-wave position coordinate systems, wherein M represents a distance dimension, and L represents a wave position dimension;
(3) selecting a distance-wave position resolution unit:
selecting points from the distance-wave position coordinate system, and forming a distance-wave position resolution unit by the distance-wave position corresponding to the points;
(4) calculating an amplitude and phase linearity matrix:
(4a) calculating the amplitude value of the selected distance-wave position resolution unit by using an amplitude value formula;
(4b) calculating the phase linearity value of the selected distance-wave position resolution unit by using a phase linearity value formula;
(4c) calculating the ratio of the amplitude value and the phase linearity value of the selected distance-wave position resolution unit by using an amplitude and phase linearity ratio formula;
(4d) calculating the elements of the amplitude and phase linearity matrix by using an amplitude and phase linearity matrix element formula;
(5) judging whether all points in the distance-wave position coordinate system are selected, if so, executing the step (6); otherwise, executing the step (3);
(6) converting the amplitude and phase linearity matrix into a gray level image:
the matrix is converted into a grayscale image using mat2gray statements in matlab2014 a;
(7) generating a binary image:
(7a) acquiring an optimal threshold value of the gray level image by using a maximum inter-class variance method;
(7b) judging whether the pixel value of any pixel point in the gray level image is larger than the optimal threshold value, if so, executing the step (7 c); otherwise, executing step (7 d);
(7c) setting the pixel value of the pixel point of which the pixel value is greater than the optimal threshold value in the gray level image as 0;
(7d) setting the pixel value of the pixel point with the pixel value less than or equal to the optimal threshold value in the gray level image as 1;
(7e) setting pixel values of all pixel points in the image to be 0 and setting the pixel values to be 1 to form a binary image;
(8) generating a sea-land clutter scene segmentation image:
and performing morphological filtering on the binary image by using 5-by-5 structural elements to obtain a final sea and land clutter scene segmentation image.
Compared with the prior art, the invention has the following advantages:
, because the invention adopts the ratio of amplitude and phase linearity as measure in the process of sea and land clutter scene image segmentation, it fully embodies the difference of sea and land clutter under the platform of motion or static coherent system, overcomes the deficiency that the prior art can not correctly segment sea and land clutter with unbalanced density distribution, and improves the segmentation precision of the image.
Secondly, in the process of segmenting the sea and land clutter scene image, the threshold segmentation is carried out on the gray level image by adopting the maximum inter-class variance threshold segmentation method, the process of obtaining the optimal threshold is self-adaptive, any parameter does not need to be set artificially, the operation speed is high, and the defect of threshold selection in the prior art is overcome, so that the optimal threshold can be obtained to obtain a more satisfactory segmentation result when the image gray level histogram has no obvious double peaks or troughs.
Thirdly, in the process of sea and land clutter scene image segmentation, morphological filtering is adopted to process the binary image after threshold segmentation, so that the connectivity of a land area and an ocean area in a segmentation result is ensured, the defect of scattered segmentation in the prior art is overcome, and the quality of sea and land clutter scene segmentation is improved.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a simulation of the present invention.
Detailed Description
The present invention is described in further detail with reference to the attached figures.
Step 1, the radar receives scatterers and returns a three-dimensional echo sequence containing pulse-distance-wave positions.
And 2, establishing a distance-wave position coordinate system.
And converting the echo sequence into a gray image, and establishing a distance-wave position coordinate system.
And 3, selecting a distance-wave position resolution unit.
points are selected from the distance-wave position coordinate system, and the distance-wave position corresponding to the points is formed into a distance-wave position resolution unit.
And 4, calculating an amplitude and phase linearity matrix.
Calculating the amplitude value of the selected distance-wave position resolution unit by using an amplitude value formula;
the amplitude value formula is as follows:
wherein, PkRepresenting the amplitude value of the kth range-wave-bit-resolution cell, i representing the pulse sequence of the echo sequence
N, N denotes the total number of pulses of the echo sequence, Σ denotes a summation operation, | · | denotes an amplitude operation, x denotes an amplitude operationi,kThe k-th range-bit resolution unit represents the i-th pulse of the echo sequence, k being 1, 2.
Calculating the phase linearity value of the selected distance-wave position resolution unit by using a phase linearity value formula;
the phase linearity value formula is as follows:
wherein, thetakRepresents the phase linearity value of the kth range-wave-bit resolution unit,denotes the square operation, N denotes the pulse number of the echo sequence, N ═ 2,32Denotes squaring operation, Δk(n) a phase difference between a winding phase and a real phase of a kth range-bit resolution cell of adjacent pulses of the echo train,the wrap phase, phi, of the kth range-phase resolution unit representing the nth pulse of the echo sequencek(n) represents the true phase of the kth range-bit resolution cell of the nth pulse of the echo sequence,phase difference delta between the winding phase and the real phase of range-wave position resolution elements representing all adjacent pulses of an echo sequencek(n) the average value of (n),
calculating the ratio of the amplitude value and the phase linearity value of the selected distance-wave position resolution unit by using an amplitude and phase linearity ratio formula;
the amplitude-phase linearity ratio formula is as follows:
wherein R iskRepresenting amplitude values P of a kth range-bit resolution cellkAnd the phase linearity value thetakThe ratio of (a) to (b).
Calculating the elements of the amplitude and phase linearity matrix by using an amplitude and phase linearity matrix element formula;
the formula of the amplitude and phase linearity matrix element is as follows:
Rpq=log10(Rk);
wherein R ispqRepresents the elements of the P-th row and Q-th column in the amplitude and phase linearity matrix, P represents the distance number from the wave-position resolution unit, P is 1,210A base 10 logarithmic operation is shown.
And 5, judging whether all points in the distance-wave position coordinate system are selected, if so, executing the step 6, and otherwise, executing the step 3.
And 6, converting the amplitude and phase linearity matrix into a gray image.
And 7, generating a binary image.
(7a) And acquiring the optimal threshold of the gray level image by using a maximum inter-class variance method.
The method for the variance between the maximum classes comprises the following specific steps:
, dividing the gray image into S gray levels by arbitrarily selecting t1And C2Two kinds, C1A set of pixels representing a gray value of t or less in a gray image, C1={1,2,...,t},C2Representing a set of pixels having a gray value greater than t in a gray image, C2={t+1,t+2,...,S-1},0≤t≤S-1;
Second, calculate C according to the following equation1And C2The inter-class variance of (c) is,
σ2(t)=p1(μ1-μT)2+p2(μ2-μT)2
wherein σ2(t) represents C1And C2Between classes of variance, p1Is represented by C1The probability of occurrence of the event is,nλrepresenting the total number of pixels with a gray value of lambda, W representing the gray imageTotal number of pixels of p2Is represented by C2The probability of occurrence of the event is,μ 1 represents C1The mean value of the gray levels of (a),μ 2 represents C2The mean value of the gray levels of (a),μ (t) represents a grayscale mean of the grayscale image,
third, calculate σ as follows2(t) a threshold value at which the maximum value is obtained,
T=Argmax{σ2(t)}
wherein T represents the between-class variance σ2(t) a threshold value at which a maximum value is obtained, Argmax representing a variable operation corresponding to the maximum value;
and fourthly, taking the threshold T as the optimal threshold of the gray-scale image.
(7b) Judging whether the pixel value of a pixel point in the gray level image is larger than the optimal threshold value, if so, executing the step (7 c); otherwise, step (7d) is performed.
(7c) And setting the pixel value of the pixel point which is greater than the optimal threshold value in the pixel values of the pixel points in the gray level image as 0.
(7d) And setting the pixel value of the pixel point with the pixel value less than or equal to the optimal threshold value in the gray level image as 1.
(7e) Setting pixel values of all pixel points in the image to be 0 and setting the pixel values to be 1 to form a binary image;
and 8, generating a sea and land clutter scene segmentation image.
The morphological filtering is to adopt 5 × 5 structural elements to perform opening and closing operations on the binary image, filter isolated clusters occupying a plurality of pixels in the binary image, fill holes in a connected region, and divide land and ocean two scene regions to obtain a final sea and land clutter scene segmentation image.
The effect of the present invention can be further illustrated by following simulations:
1. simulation conditions are as follows:
the simulation experiment of the invention is carried out in the environment that the computer is configured with core i73.40GHZ, the memory 8G, the WINDOWS7 system and the computer software are configured with Matlab R2014 a.
2. Simulation content:
the simulation experiment of the invention uses two prior arts (a phase linearity method, a standard fuzzy C-means method) and the invention to respectively simulate a pulse-distance-wave position three-dimensional echo sequence which is received by a radar and contains a sea and land clutter scene, and the obtained results are shown in fig. 2(b), fig. 2(C) and fig. 2 (d).
FIG. 2(a) is an original image obtained by calculating the amplitude of a pulse-distance-wave position three-dimensional echo sequence of a scene containing sea and land clutter received by a radar and converting the amplitude;
fig. 2(b) is a simulation graph obtained by using a phase linearity method in the prior art, where fig. 2(b) is a sea and land clutter scene segmentation image obtained by calculating phase linearity of all distance-wave position resolution units in an echo sequence by using the phase linearity method, converting the phase linearity into a gray image, performing threshold processing on the gray image by using a maximum inter-class variance method to obtain a binary image, and performing morphological filtering on the binary image by using 8 × 8 structural elements;
FIG. 2(C) is a simulation diagram obtained by using a standard fuzzy C-means method in the prior art, wherein FIG. 2(C) is a sea and land clutter scene segmentation image obtained by calculating a clustering center and a membership matrix in an image transformed by an echo sequence by using the standard fuzzy C-means method and generating an optimal C interval by minimizing a weighted clustering objective function;
fig. 2(d) is a simulation graph obtained by using the method of the present invention, where fig. 2(d) is a sea and land clutter scene segmentation image obtained by calculating an amplitude and phase linearity matrix of a distance-wave position resolution unit in an echo sequence by using the method of the present invention, converting the matrix into a gray image, performing threshold processing on the gray image by using a maximum inter-class variance method to obtain a binary image, and performing morphological filtering on the binary image by using 5 × 5 structural elements;
3. and (3) simulation result analysis:
as can be seen from fig. 2(b) and 2(d), the clutter scene segmentation method based on phase linearity in the prior art leads to many holes to be reserved being connected, resulting in under-segmentation and poor segmentation quality. The invention fully reflects the difference of sea and land clutter and can obtain more accurate segmentation results.
As can be seen from fig. 2(c) and 2(d), the image segmentation method using FCM clustering in the prior art cannot correctly segment sea and land clutter with unbalanced density distribution, and the segmentation quality is poor. The method can better divide the sea and land clutter with unbalanced density distribution and can obtain more accurate division results.
Claims (7)
1, A sea and land clutter scene segmentation method based on amplitude and phase linearity, comprising the following steps:
(1) acquiring an echo sequence:
receiving a pulse-distance-wave position three-dimensional echo sequence containing a sea and land clutter scene by a radar;
(2) establishing a distance-wave position coordinate system:
converting the echo sequence into a grayscale image with the size of M multiplied by L, and establishing distance-wave position coordinate systems, wherein M represents a distance dimension, and L represents a wave position dimension;
(3) selecting a distance-wave position resolution unit:
selecting points from the distance-wave position coordinate system, and forming a distance-wave position resolution unit by the distance-wave position corresponding to the points;
(4) calculating the ratio of amplitude to phase linearity:
(4a) calculating the amplitude value of the selected distance-wave position resolution unit by using an amplitude value formula;
(4b) calculating the phase linearity value of the selected distance-wave position resolution unit by using a phase linearity value formula;
(4c) calculating the ratio of the amplitude value and the phase linearity value of the selected distance-wave position resolution unit by using an amplitude and phase linearity ratio formula;
(4d) calculating the elements of the amplitude and phase linearity matrix by using an amplitude and phase linearity matrix element formula;
(5) judging whether all points in the distance-wave position coordinate system are selected, if so, executing the step (6);
otherwise, executing the step (3);
(6) converting the amplitude and phase linearity matrix into a gray level image;
(7) generating a binary image:
(7a) acquiring an optimal threshold value of the gray level image by using a maximum inter-class variance method;
(7b) judging whether the pixel value of a pixel point in the gray level image is larger than the optimal threshold value, if so, executing the step (7 c); otherwise, executing step (7 d);
(7c) setting the pixel value of a pixel point which is greater than the optimal threshold value in the pixel values of the pixel points in the gray level image as 0;
(7d) setting the pixel value of the pixel point with the pixel value less than or equal to the optimal threshold value in the gray level image as 1;
(7e) setting pixel values of all pixel points in the image to be 0 and setting the pixel values to be 1 to form a binary image;
(8) generating a sea-land clutter scene segmentation image:
and performing morphological filtering on the binary image by using 5-by-5 structural elements to obtain a final sea and land clutter scene segmentation image.
2. The method of claim 1, wherein the method comprises: the amplitude value formula in step (4a) is as follows:
wherein, PkRepresents the amplitude value of the kth range-wave position resolution unit, i represents the pulse number of the echo train, i is 1,2The total number of pulses of a column, Σ represents a summation operation, | · | represents an amplitude operation, xi,kThe k-th range-bit resolution unit represents the i-th pulse of the echo sequence, k being 1, 2.
3. The method of claim 1, wherein the method comprises: the phase linearity value formula in step (4b) is as follows:
wherein, thetakRepresents the phase linearity value of the kth range-wave-bit resolution unit,represents an operation on the start side, N represents the total number of pulses of the echo sequence, N represents the pulse number of the echo sequence, N is 2,32Denotes squaring operation, Δk(n) a phase difference between a winding phase and a real phase of a kth range-bit resolution cell of adjacent pulses of the echo train, the wrap phase, phi, of the kth range-phase resolution unit representing the nth pulse of the echo sequencek(n) represents the true phase of the kth range-bit resolution cell of the nth pulse of the echo sequence,phase difference delta between the winding phase and the real phase of range-wave position resolution elements representing all adjacent pulses of an echo sequencek(n) the average value of (n),
4. the method of claim 1, wherein the method comprises: the amplitude-to-phase linearity ratio formula in step (4c) is as follows:
wherein R iskRepresenting amplitude values P of a kth range-bit resolution cellkAnd the phase linearity value thetakN denotes the total number of pulses of the echo sequence, xi,kThe kth range-wave-position resolution unit, delta, representing the ith pulse of an echo sequencek(n) a phase difference between a winding phase and a real phase of a kth range-bit resolution cell of adjacent pulses of the echo train, the wrap phase, phi, of the kth range-phase resolution unit representing the nth pulse of the echo sequencek(n) represents the true phase of the kth range-bit resolution cell of the nth pulse of the echo sequence,phase difference delta between the winding phase and the real phase of range-wave position resolution elements representing all adjacent pulses of an echo sequencek(n) the average value of (n),
5. the method of claim 1, wherein the method comprises: the formula of the amplitude and phase linearity matrix element in step (4d) is as follows:
Rpq=log10(Rk)
wherein R ispqRepresents the elements of the P-th row and Q-th column in the amplitude and phase linearity matrix, P represents the distance number from the wave position resolution unit, P is 1,210Representing base 10 logarithmic operation, RkRepresenting amplitude values P of a kth range-bit resolution cellkAnd the phase linearity value thetakRatio of (A) to (B), PkRepresenting the amplitude value, theta, of the kth range-bit resolution cellkRepresenting the phase linearity value of the kth range-bit resolution unit.
6. The method of claim 1, wherein the method comprises: the method for the variance between the maximum classes in the step (7a) comprises the following specific steps:
, dividing the gray image into S gray levels by arbitrarily selecting t1And C2Two kinds, C1A set of pixels representing a gray value of t or less in a gray image, C1={1,2,...,t},C2Representing a set of pixels having a gray value greater than t in a gray image, C2={t+1,t+2,...,S-1},0≤t≤S-1;
Second, calculate C according to the following equation1And C2Inter-class variance of (c):
σ2(t)=p1(μ1-μT)2+p2(μ2-μT)2
wherein σ2(t) represents C1And C2Between classes of variance, p1Is represented by C1The probability of occurrence of the event is,nλrepresenting a total number of pixels with a grey value of lambdaNumber, W, represents the total number of pixels of the gray scale image, p2Is represented by C2The probability of occurrence of the event is,μ1is represented by C1The mean value of the gray levels of (a),μ2is represented by C2The mean value of the gray levels of (a),μTa mean value of the gray levels representing the gray-level image,
third, calculate σ as follows2(t) threshold value at which maximum value is obtained:
T=Arg max{σ2(t)}
wherein T represents the between-class variance σ2(t) a threshold value at which a maximum value is obtained, Argmax representing a variable operation corresponding to the maximum value;
and fourthly, taking the threshold T as the optimal threshold of the gray-scale image.
7. The method of claim 1, wherein the method comprises: the morphological filtering in the step (8) is to perform opening and closing operations on the binary image by using 5 × 5 structural elements, filter isolated clusters occupying a plurality of pixels in the binary image, fill holes in a connected region, divide land and ocean two scene regions, and obtain a final sea and land clutter scene segmentation image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710720482.1A CN107610130B (en) | 2017-08-21 | 2017-08-21 | Sea-land clutter scene segmentation method based on amplitude and phase linearity ratio |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710720482.1A CN107610130B (en) | 2017-08-21 | 2017-08-21 | Sea-land clutter scene segmentation method based on amplitude and phase linearity ratio |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107610130A CN107610130A (en) | 2018-01-19 |
CN107610130B true CN107610130B (en) | 2020-01-31 |
Family
ID=61065625
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710720482.1A Active CN107610130B (en) | 2017-08-21 | 2017-08-21 | Sea-land clutter scene segmentation method based on amplitude and phase linearity ratio |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107610130B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108961255B (en) * | 2018-06-28 | 2021-09-28 | 西安电子科技大学 | Sea-land noise scene segmentation method based on phase linearity and power |
CN109543589B (en) * | 2018-11-16 | 2021-02-02 | 西安电子科技大学 | Sea-land scene segmentation method based on initial phase-Doppler invariant distance and KNN |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104239901A (en) * | 2014-09-11 | 2014-12-24 | 西安电子科技大学 | Polarized SAR image classification method based on fuzzy particle swarm and target decomposition |
CN105427301A (en) * | 2015-11-17 | 2016-03-23 | 西安电子科技大学 | Sea and land clutter scene segmentation method based on direct current component ratio measure |
CN105844644A (en) * | 2016-03-31 | 2016-08-10 | 西安电子科技大学 | Morphological median derivative-based sea-land clutter scene segmentation method |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8515167B2 (en) * | 2009-08-31 | 2013-08-20 | Peking University | High dynamic range image mapping with empirical mode decomposition |
-
2017
- 2017-08-21 CN CN201710720482.1A patent/CN107610130B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104239901A (en) * | 2014-09-11 | 2014-12-24 | 西安电子科技大学 | Polarized SAR image classification method based on fuzzy particle swarm and target decomposition |
CN105427301A (en) * | 2015-11-17 | 2016-03-23 | 西安电子科技大学 | Sea and land clutter scene segmentation method based on direct current component ratio measure |
CN105844644A (en) * | 2016-03-31 | 2016-08-10 | 西安电子科技大学 | Morphological median derivative-based sea-land clutter scene segmentation method |
Non-Patent Citations (1)
Title |
---|
"陆海杂波场景快速分割方法研究";马晓礼;《中国优秀硕士学位论文全文数据库 信息科技辑》;20160315;第1-5章 * |
Also Published As
Publication number | Publication date |
---|---|
CN107610130A (en) | 2018-01-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ao et al. | Detection and discrimination of ship targets in complex background from spaceborne ALOS-2 SAR images | |
CN108961255B (en) | Sea-land noise scene segmentation method based on phase linearity and power | |
CN105427301B (en) | Based on DC component than the extra large land clutter Scene Segmentation estimated | |
CN109116352A (en) | A kind of circle sweeps ISAR mode ship super-resolution imaging method | |
CN101482969B (en) | SAR image speckle filtering method based on identical particle computation | |
CN104331886A (en) | Port region ship and warship detection method based on high resolution SAR image | |
CN113362293A (en) | SAR image ship target rapid detection method based on significance | |
CN105741262B (en) | The extra large land clutter Scene Segmentation estimated based on energy compaction measure | |
CN106443593A (en) | Self-adaptive oil spill information extraction method based on coherent radar slow-scan enhancement | |
CN107610130B (en) | Sea-land clutter scene segmentation method based on amplitude and phase linearity ratio | |
CN111812598A (en) | Time domain and frequency domain multi-feature-based ground and sea clutter classification method | |
Celona et al. | Automated detection, classification, and tracking of internal wave signatures using X-band radar in the inner shelf | |
CN109543589B (en) | Sea-land scene segmentation method based on initial phase-Doppler invariant distance and KNN | |
CN107909595A (en) | Extra large land clutter Scene Segmentation based on amplitude Yu energy compaction measure product | |
CN112101250B (en) | Method for detecting offshore ship target based on context semantic perception | |
CN105844644B (en) | Extra large land clutter Scene Segmentation based on morphology intermediate value derivative | |
CN112215832B (en) | SAR trail image quality evaluation and self-adaptive detection parameter adjustment method | |
CN112435249A (en) | Dynamic small target detection method based on periodic scanning infrared search system | |
CN111583267A (en) | Generalized fuzzy C-means clustering-based fast SAR image sidelobe suppression method | |
CN114373135A (en) | Ship target detection method based on local significance characteristic measurement | |
CN115294398A (en) | SAR image target recognition method based on multi-attitude angle joint learning | |
Chen et al. | Integrated processing of radar detection and classification for moving target via time-frequency graph and CNN learning | |
CN113820680A (en) | Multi-frame sea-land radar echo segmentation method based on covariance | |
CN112816956A (en) | Method and device for acquiring radar target information | |
He et al. | Research on Solid Rate Filtering Technique based on Inverse Distance Weighted Interpolation of Navigation Radar |
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 |