CN110728695B - Video SAR moving target detection method based on image area accumulation - Google Patents

Video SAR moving target detection method based on image area accumulation Download PDF

Info

Publication number
CN110728695B
CN110728695B CN201911003255.2A CN201911003255A CN110728695B CN 110728695 B CN110728695 B CN 110728695B CN 201911003255 A CN201911003255 A CN 201911003255A CN 110728695 B CN110728695 B CN 110728695B
Authority
CN
China
Prior art keywords
accumulation
image
window
size
target
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
Application number
CN201911003255.2A
Other languages
Chinese (zh)
Other versions
CN110728695A (en
Inventor
丁金闪
仲超
徐众
柯凌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xidian University
Original Assignee
Xidian University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Xidian University filed Critical Xidian University
Priority to CN201911003255.2A priority Critical patent/CN110728695B/en
Publication of CN110728695A publication Critical patent/CN110728695A/en
Application granted granted Critical
Publication of CN110728695B publication Critical patent/CN110728695B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/215—Motion-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10016—Video; Image sequence
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10032—Satellite or aerial image; Remote sensing
    • G06T2207/10044—Radar image

Abstract

The invention discloses a video SAR moving target detection method based on image region accumulation, which mainly solves the problem that the existing video SAR moving target detection is not stable. The implementation scheme is as follows: 1) designing an accumulation window and carrying out regional accumulation on the video SAR image sequence; 2) determining an accumulation threshold and carrying out image reformation on an accumulation result; 3) performing binary segmentation on the reformed image; 4) carrying out connected domain size statistics on the image after binary segmentation and reserving a connected domain with the size within the range of 0.4 time of the total number of the pixels occupied by the target and 2 times of the total number of the pixels occupied by the target; 5) and performing inter-frame correlation processing on the image subjected to connected domain processing to remove non-target shadows and finish moving target detection. The invention effectively inhibits false alarm and false alarm missing probability and improves detection performance by carrying out regional multi-frame joint detection on the video SAR image, and can be used for tracking and detecting a moving target of the video SAR in real time.

Description

Video SAR moving target detection method based on image area accumulation
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a SAR moving target detection method which can be used for tracking and detecting a video SAR moving target in real time.
Background
The video SAR imaging radar can perform real-time imaging on a ground target area all day long, all weather and high precision, has the characteristics of high imaging frame rate and high resolution, can effectively overcome the defect that an infrared/visible light sensor is easily influenced by weather conditions and battlefield environments, and can overcome the defects of low frame rate, difficulty in moving target detection and tracking and the like of a conventional SAR system. When the video SAR imaging is carried out, the image of a moving target can be defocused and shifted, but shadow can be left at the real position of the image, and the imaging shadow can be moved in the image sequence obtained in the video imaging mode. Therefore, the dynamic shadow-based data processing technology can realize the detection and the tracking of the moving target in principle. However, since the SAR uses coherent electromagnetic waves for imaging, a large amount of speckle noise exists in the SAR image, and a large amount of false alarms are generated when the target is detected by using shadows, it is of great significance to perform robust detection on a moving target in the video SAR.
Most of the existing moving target detection methods based on dynamic shadows in video SAR are based on the image registration, and change information in an image sequence is obtained by extracting an image sequence of a static background and subtracting the image sequence of a scene video SAR. The method is based on the operation of a single pixel point, and has the problems of inaccurate background extraction and overhigh false alarm rate. In recent years, more and more detection methods are applied to the moving target detection of the video SAR, and therefore, a moving target detection technology based on the video SAR image frame is also developed.
Doerry A, Miller J, Bishop E et al in the article "Shadow Proavailability of Detection and False Alarm for media-Filtered SAR image" suppress speckle noise and improve the single-point Detection performance of the Shadow region in the SAR image by a Median filtering method. In the article "a novel approach to moving targets detection in video image sequence", Y Zhang, X Mao, etc., dynamic shadows are extracted by image processing for video data published in Sandia laboratories, thereby achieving the purpose of moving target detection. Both methods are based on the operation of image pixels, and because the relationship between pixels of the shadow of the moving target and the time sequence change relationship of the shadow between frames are not considered, the problems of over-high false alarm and over-high false alarm are easy to occur, and the detection performance of the low-speed target and the high-maneuvering target point is deteriorated.
Disclosure of Invention
The invention aims to provide a video SAR moving target detection method based on image area accumulation aiming at the defects of the prior art, so as to reduce false alarm and false-miss alarm influence caused by interframe shadow time sequence change in a video SAR and improve the detection probability of a moving target.
The technical scheme for realizing the aim of the invention comprises the following steps:
(1) carrying out regional accumulation on a video SAR image sequence:
1a) setting an accumulation window, and determining the size of the accumulation window to be N x N according to the resolution of the video SAR image sequence and the size of the shadow of the moving target in the image, wherein N is an odd number, and x represents dot multiplication;
1b) accumulating image areas by using a sliding accumulation window method, wherein the center of the accumulation window moves point by point from a first image pixel, and the value of each corresponding pixel point is determined by summing the image intensity values covered in the accumulation window around the point;
1c) repeating the step 1b) until the center of the accumulation window completely traverses the video SAR image sequence to generate an accumulated image sequence;
(2) determining an accumulation threshold STAnd performing image reformation on the accumulated result:
2a) according to a uniform background probability density function fB(s), system noise level λ and false alarm rate PfaUsing the formulaTNamely:
Figure BDA0002241960750000021
2b) using an accumulation threshold STUniformly reforming the accumulated image sequence generated in 1c), namely setting a reforming frame with the same size as the accumulation frame in 1a), continuously moving the center of the reforming window from the first pixel in the accumulated image, and if the energy of the central pixel in the reforming frame is lower than the accumulation threshold S determined in 2a)TIf not, adding 0 to all the pixel points covered in the reforming frame;
2c) repeating the moving and accumulation operations of the center of the reforming window in the step 2b) until the center of the reforming window completely traverses the image sequence, so that each pixel point completes N times of accumulation operation, and the value range of the pixel point is [0, N times N ];
(3) selecting N x (N-1)/2 as a threshold for binary segmentation of the reformed image, wherein the value range of the segmented image is [0,1 ];
(4) performing connected domain statistics on the image subjected to binary segmentation in the step (3), and deleting large connected domains with the number of pixel points being more than 2 times of the total number of the pixel points occupied by the target and small connected domains with the number of the pixel points being less than 0.4 time of the total number of the pixel points occupied by the target in the image;
(5) and (5) performing inter-frame correlation processing on the image processed in the step (4) to finish detection of a moving target:
5a) initializing a moving target detection set, namely taking all shadow detection results in an initial frame as initial moving target shadows;
5b) retaining the shadow of which the displacement between adjacent frames exceeds half of the size of the moving target in the detection result of the previous frame by frame, and deleting the shadow of which the displacement between adjacent frames does not exceed half of the size of the moving target;
5c) and deleting the static shadows continuously appearing in the 3 frames of images in the shadows reserved in the step 5b) and the flicker shadows randomly appearing in the 3 frames of images in the continuous 5 frames of images, and finally obtaining the detection result of the moving target.
Compared with the prior art, the invention has the following advantages:
1. according to the method, the video SAR image sequence is scanned and detected pixel by pixel, the joint area information of each pixel is judged, and the influence of coherent speckles on detection is effectively inhibited.
2. The invention greatly improves the detection performance of the shadow area by carrying out secondary threshold segmentation on the accumulated image.
3. The invention avoids the extraction of the background by utilizing the tracking thought, thereby greatly reducing the calculation brought by the image registration and the edge error brought by the background extraction.
4. The invention relates to the dynamic information among multiple frames, effectively removes static shadows and flickering false moving targets, and improves the detection accuracy of real moving targets.
Drawings
FIG. 1 is a flow chart of an implementation of the present invention;
FIG. 2 is a schematic illustration of the region accumulation of an image through an accumulation window in the present invention;
FIG. 3 is a schematic diagram of the present invention illustrating the unified reformation of an image;
fig. 4 is a simulation result diagram of video SAR moving target detection using the present invention.
Detailed Description
Embodiments and effects of the present invention will be further described below with reference to the accompanying drawings.
Referring to fig. 1, the implementation steps of the present invention are as follows:
designing an accumulation window and carrying out regional accumulation on the high frame rate video SAR image sequence.
1.1) setting the accumulation window size N:
the size of the accumulation window is determined by the resolution of the video SAR image sequence and the size of the moving object in the image, a too small window size cannot effectively accumulate shadow areas, and a too large window size deteriorates the resolution of the image sequence. The design rule of the accumulation window is to be equal to the size of the shadow of the moving object and not larger than the size of the shadow of the moving object so as to ensure that a part of shadow area can be completely covered by the accumulation window, and the square accumulation window is adopted in consideration of the fact that the moving direction of the moving object cannot be obtained in a priori.
In this example, as shown in fig. 4(a), the 104 th frame is selected from the video SAR image sequence, the resolution of the SAR image is 0.2m, and the size of the moving target is 1.8m × 4.6m, so that the shadow of the moving target will occupy at least 9 × 23 pixels in the SAR image, and to ensure that the shadow region is effectively accumulated, N ═ 2 × ceil (min (N) ("N")1,N2) /2) -1, where ceil () represents rounded up, a dot product, N1,N2The size of a two-dimensional pixel point occupied by a moving target in an image is obtained;
in this case N1＝9,N2If 23, taking N to 7, the size of the accumulation window is designed to 7 by 7;
1.2) image area accumulation is carried out by using a sliding window method, and the intensity of the center of a current accumulation window is obtained:
as shown in fig. 2, in this step, the accumulation window is used to perform area accumulation on a new image of the video SAR by traversing pixel points one by one, that is, the center of the accumulation window is continuously moved from a first image pixel, so that the accumulation window covers different image areas in the moving process, and the intensity of each pixel point in the accumulation window is changed accordingly; and summing the intensities of all pixel points in the area covered by the accumulation window to obtain the intensity of the center of the current accumulation window.
1.3) repeating the operation of 1.2) until the center of the accumulation window completely traverses the video SAR image sequence to obtain the accumulated image sequence.
The accumulated image is shown in fig. 4 (b).
Step two, determining an accumulation threshold ST。
The accumulation intensity in the accumulation window is comprehensively determined by the background intensity, the noise level and the shadow region intensity in the SAR image, and the image intensity threshold S after region accumulation can be derived according to the requirements of the false alarm rate and the detection rateT。
2.1) for a uniform background region, each pixel point IiSubject to an independent and identically distributed exponential distribution I of the intensity values of the imagei～exp(λi) The probability density function is:
Figure BDA0002241960750000041
wherein λi＝1/σiRepresenting the overall equivalent backscattering coefficient σ of the ith ground resolution elementiThe reciprocal of (a);
2.2) selecting 7 × 7 resolution cells in the accumulation window, the accumulation intensity S of each pixel point in the accumulation window is expressed as:
Figure BDA0002241960750000042
2.3) dividing the 49 resolution cell intensities S within the accumulation window into n groups, each group having the same parameter λiThe probability density function for calculating the accumulation intensity is:
Figure BDA0002241960750000051
wherein λi,piRespectively representing the parameters of the exponential distribution and the number of random variables corresponding to the parameters, mlTo conform to m1+...+mn＝pi-jAnd miAll natural numbers not equal to 0 condition, l ═ 1,2.. n;
2.4) calculation in<4>Probability of detection P under the condition of probability density functiondComprises the following steps:
Figure BDA0002241960750000052
2.5) calculating the probability density function of the accumulated intensity obeying the Gamma distribution when the background is a uniform area according to the statistical principle:
Figure BDA0002241960750000053
wherein Γ (N) represents a gamma function with a parameter of accumulation window size N;
2.6) given accumulation threshold STCalculating the false alarm probability of the shadow area:
Figure BDA0002241960750000054
2.7) utilization formula<5>A and B type<7>And a background scattering intensity parameter λiAt a detection probability P of not less than 90%dAnd a false alarm rate P of not more than 1%faUnder the condition of (1), selecting an accumulation threshold ST。
In this example, the accumulation threshold is chosen to be 5.
And step three, performing regional image reforming.
3.1) setting a reforming frame with the same size as the accumulation frame in the step 1.1), continuously moving the center of the reforming window from the first pixel in the accumulated image, and after each movement, if the total energy of the image of the center pixel of the reforming frame is lower than the accumulation threshold S determined in the step twoTIf so, performing an add-1 operation on all the pixel points covered in the reforming frame, otherwise, performing an add-0 operation, as shown in fig. 3;
3.2) repeating the operation of 3.1) until the center of the reforming window completely traverses the accumulated image sequence, after the reforming of the whole image is completed, each pixel completes 7 × 7 accumulation operations, the value range is [0,49], and the reformed image is as shown in fig. 4 (c).
And step four, performing binary segmentation on the image reformed in the step three.
And selecting N x (N-1)/2 as a threshold for the reformed image, performing 1 assigning operation on pixel units with pixel values larger than or equal to the threshold, and performing 0 assigning operation on pixel units with pixel values smaller than the threshold to finish binary segmentation.
In this example, the segmentation threshold is set to 21, and the segmentation result is shown in fig. 4 (d). As can be seen from fig. 4(d), the segmented image has a fragmented low intensity region and some fixed shadows due to the static object height. To reduce the amount of computation in subsequent steps, attempts should be made to reduce the number of such shadows produced by non-moving objects.
And step five, performing connected domain shadow statistics on the picture after binary segmentation and processing the picture.
The connected domain is divided into a single connected domain and a multi-connected domain, and the definition of the single connected domain is as follows: for an area G on the complex plane, a simple closed curve is made in the area, if the interior of the closed curve completely belongs to the area G, the area G is called a single connected domain, and any area which is not the single connected domain is a multi-connected domain.
In the fourth step, most of the connected domains in the segmented image are single connected domains, and no multi-connected domain appears, as shown in fig. 4(d) of the present example, but in some very special cases, multi-connected domains appear in the segmented image, and since no multi-connected domain is generated by a moving target, the multi-connected domains can be eliminated in the subsequent inter-frame correlation processing, only the size of the connected domain is concerned in the step, and the single connected domain and the multi-connected domain are not distinguished.
The specific implementation of the step is as follows: firstly, finding out all connected domains, taking the total number of pixel points occupied by each connected domain in the image as the size of the current connected domain, and then, according to the total number N of pixel points occupied by the target in the image1*N2Retention size at [0.4 x N1*N2,2*N1*N2]Connected domains within the range.
In this example, the number of resolution units occupied by the target on the image is 200, so that a connected domain with the number of pixels [80,400] is selected as a candidate moving target shadow region, and the detection result after the selection is shown in fig. 4 (e).
And step six, performing inter-frame correlation processing on the image subjected to connected domain processing to remove non-target shadows and finish moving target detection.
After the screening of the size of the connected domain shadow is finished, a fixed shadow with the size similar to that of the moving target shadow and the shape similar to that of the moving target shadow is also remained, and the shadow cannot be judged whether to be the moving target shadow from a single frame image, so that a static shadow and the moving target shadow need to be distinguished through the relationship among a plurality of frames of images, and the static shadow needs to be removed. Furthermore, during the screening process, the number of pixels of partial shadow is at the edge of [80,400], so that in some image frames, the area can be determined as the shadow of the moving object, unlike the shadow of the moving object, such fixed shadow appears intermittently at the same position in the observation sequence, and the phenomenon of shadow flicker is generated in the continuous frames, and the phenomenon of flicker shadow should be eliminated.
The existing methods for eliminating non-target shadows generally include a background subtraction method, an optical flow method and a track correlation method. The embodiment adopts but is not limited to a track association method, and the specific implementation steps are as follows:
6.1) initializing a moving target detection set, namely taking all shadow detection results in an initial frame as initial moving target shadows;
6.2) retaining the shadow of which the displacement between adjacent frames in the detection result of the previous frame does not exceed half of the target size frame by frame, and deleting the shadow of which the displacement between adjacent frames exceeds half of the target size;
6.3) deleting the static shadows continuously appearing in the 3 frames of images in the shadows reserved in the 6.2) and the flicker shadows randomly appearing in the 3 frames of images in the continuous 5 frames of images to finally obtain the detection result of the moving object, as shown by a white box in a figure 4 (g).
The foregoing description is only a specific example of the present invention and is not intended to limit the invention, so that it will be apparent to those skilled in the art that various changes and modifications in form and detail may be made therein without departing from the spirit and scope of the invention.

Claims (5)

1. A video SAR moving target detection method based on image region accumulation is characterized by comprising the following steps:
(1) carrying out regional accumulation on a video SAR image sequence:
1a) setting an accumulation window, and determining the size of the accumulation window to be N x N according to the resolution of the video SAR image sequence and the size of the shadow of the moving target in the image, wherein N is an odd number and x represents dot multiplication;
1b) accumulating the image area by using a sliding accumulation window method, wherein the center of the accumulation window continuously moves from a first image pixel, and the value of each corresponding pixel point is determined by summing the image sequence values covered in the accumulation window around the point;
1c) repeating the step 1b) until the center of the accumulation window completely traverses the video SAR image sequence to generate an accumulated image sequence;
(2) determining an accumulation threshold STAnd performing image reformation on the accumulated result:
2a) according to a uniform background probability density function fB(s), system noise level λ and false alarm rate PfaAnd a detection probability PdIn combination with the following equation, determining an accumulation threshold STNamely:
Figure FDA0003381398890000011
wherein s represents the cumulative intensity;
2b) using an accumulation threshold STUniformly reforming the accumulated image sequence generated in 1c), namely setting a reforming window with the same size as the accumulation frame in 1a), continuously moving the center of the reforming window from the first pixel in the accumulated image, and if the center of the reforming window is imagedThe energy of the element is lower than the accumulation threshold S determined in 2a)TIf not, adding 0 to all the pixel points covered in the reforming window by 1;
2c) repeating the moving and accumulation operations of the center of the reforming window in the step 2b) until the center of the reforming window completely traverses the image sequence, so that each pixel point completes N times of accumulation operation, and the value range of the pixel point is [0, N times N ];
(3) selecting N x (N-1)/2 as a threshold for binary segmentation of the reformed image, wherein the value range of the segmented image is [0,1 ];
(4) performing connected domain statistics on the image subjected to binary segmentation in the step (3), and deleting large connected domains with the number of pixel points being more than 2 times of the total number of the pixel points occupied by the target and small connected domains with the number of the pixel points being less than 0.4 time of the total number of the pixel points occupied by the target in the image;
(5) and (5) performing inter-frame correlation processing on the image processed in the step (4) to finish detection of a moving target:
5a) initializing a moving target detection set, namely taking all shadow detection results in an initial frame as initial moving target shadows;
5b) retaining the shadow of which the displacement between adjacent frames in the detection result of the previous frame does not exceed half of the target size frame by frame, and deleting the shadow of which the displacement between adjacent frames exceeds half of the target size;
5c) and deleting the static shadows continuously appearing in the 3 frames of images in the shadows reserved in the step 5b) and the flicker shadows randomly appearing in the 3 frames of images in the continuous 5 frames of images, and finally obtaining the detection result of the moving target.
2. The method of claim 1, wherein the determination of the size of the accumulation window in step 1a) as a function of the resolution of the sequence of video SAR images and of the size of the shadow of the moving object in the images is based on the resolution determined by the radar system and the N occupied by the moving object in the images1*N2A resolution unit for determining the size of the accumulation window as N × N, wherein N is 2 × ceil (min (N)1,N2) /2) -1, ceil () represents rounding up.
3. The method according to claim 1, wherein the step 1b) of using a sliding accumulation window method to accumulate the image areas comprises moving the center of the accumulation window from the first image pixel continuously, so that the accumulation window covers different image areas during the moving process, resulting in the change of the intensity of each pixel point in the accumulation window, and then summing the intensities of all pixel points in the area covered by the accumulation window to obtain the intensity of the center of the current accumulation window.
4. The method according to claim 1, wherein the accumulation threshold is determined in step 2a) as follows:
2a1) calculating the intensity I of each resolution unit of the accumulation window according to the background intensity, the noise level and the shadow region intensity in the SAR imageiProbability density function of (1):
Figure FDA0003381398890000021
wherein λi＝1/σiRepresenting the overall equivalent backscattering coefficient σ of the ith ground resolution elementiThe reciprocal of (a);
2a2) selecting N × N resolution cells within the accumulation window, the accumulation intensity can be expressed as:
Figure FDA0003381398890000022
2a3) dividing the N ^2 resolution cell intensities within the accumulation window into N groups, each group having the same parameter lambdaiThe probability density function for calculating the accumulation intensity is:
Figure FDA0003381398890000031
wherein λi,piNumber of parameters, m, respectively representing the exponential distribution and the random variables of the parameterslAll conform to m1+…+mn＝pi-j and miA natural number not equal to 0 condition, l being 1,2.. n;
2a4) the target detection probability under the condition of the probability density function of the formula <4> is:
Figure FDA0003381398890000032
2a5) according to the statistical principle, the accumulated intensity obeying the Gamma distribution when the background is a uniform area is calculated, and the probability density function is:
Figure FDA0003381398890000033
wherein Γ (N) is a gamma function with a parameter N;
2a6) given accumulation threshold STThe false alarm probability of the shadow region is:
Figure FDA0003381398890000034
by using the above formula<5>、<7>And a background scattering intensity parameter λiAnd defining a detection probability P not less than 90%dAnd a false alarm rate P of not more than 1%faUnder the condition of (1), selecting an accumulation threshold ST。
5. The method according to claim 1, wherein the step (4) of performing connected domain statistics on the binary-segmented picture, wherein the total number of pixels occupied by each connected domain in the image is taken as the size of the current connected domain, and the total number of pixels N occupied by the target in the image is taken as the size of the current connected domain1*N2With a size of the connected component of [ 0.4N ]1*N2,2*N1*N2]Of the connected domain.
CN201911003255.2A 2019-10-22 2019-10-22 Video SAR moving target detection method based on image area accumulation Active CN110728695B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911003255.2A CN110728695B (en) 2019-10-22 2019-10-22 Video SAR moving target detection method based on image area accumulation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911003255.2A CN110728695B (en) 2019-10-22 2019-10-22 Video SAR moving target detection method based on image area accumulation

Publications (2)

Publication Number Publication Date
CN110728695A CN110728695A (en) 2020-01-24
CN110728695B true CN110728695B (en) 2022-03-04

Family

ID=69220558

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911003255.2A Active CN110728695B (en) 2019-10-22 2019-10-22 Video SAR moving target detection method based on image area accumulation

Country Status (1)

Country Link
CN (1) CN110728695B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108254756A (en) * 2017-12-13 2018-07-06 北京空间机电研究所 A kind of satellite-bone laser radar no-coherence cumulating detection method based on projection convolution
CN108734111A (en) * 2018-04-26 2018-11-02 西南电子技术研究所（中国电子科技集团公司第十研究所） SAR image surface vessel recognition methods
CN109031277A (en) * 2018-06-21 2018-12-18 电子科技大学 A kind of through-wall radar multi-Target Image domain robust tracking method
CN109917378A (en) * 2018-12-26 2019-06-21 西安电子科技大学 Utilize the VideoSAR moving target detecting method of space time correlation

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10275691B2 (en) * 2017-08-22 2019-04-30 Northrop Grumman Systems Corporation Adaptive real-time detection and examination network (ARDEN)

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108254756A (en) * 2017-12-13 2018-07-06 北京空间机电研究所 A kind of satellite-bone laser radar no-coherence cumulating detection method based on projection convolution
CN108734111A (en) * 2018-04-26 2018-11-02 西南电子技术研究所（中国电子科技集团公司第十研究所） SAR image surface vessel recognition methods
CN109031277A (en) * 2018-06-21 2018-12-18 电子科技大学 A kind of through-wall radar multi-Target Image domain robust tracking method
CN109917378A (en) * 2018-12-26 2019-06-21 西安电子科技大学 Utilize the VideoSAR moving target detecting method of space time correlation

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
雷达低可观测目标探测技术;陈小龙;《科技导报》;20171231;第35卷(第11期);第30-37页 *

Also Published As

Publication number Publication date
CN110728695A (en) 2020-01-24

Similar Documents

Publication Publication Date Title
JP6759475B2 (en) Ship detection methods and systems based on multidimensional features of the scene
CN109816641B (en) Multi-scale morphological fusion-based weighted local entropy infrared small target detection method
CN108182690B (en) A kind of infrared Weak target detecting method based on prospect weighting local contrast
CN110782477A (en) Moving target rapid detection method based on sequence image and computer vision system
CN111144213B (en) Object detection method and related equipment
Huang et al. Motion detection with pyramid structure of background model for intelligent surveillance systems
Sengar et al. Motion detection using block based bi-directional optical flow method
Lian et al. A novel method on moving-objects detection based on background subtraction and three frames differencing
CN108665420B (en) Infrared dim target image background suppression method based on variational Bayesian model
Ravanfar et al. Low contrast sperm detection and tracking by watershed algorithm and particle filter
CN110349188B (en) Multi-target tracking method, device and storage medium based on TSK fuzzy model
CN110060278B (en) Method and device for detecting moving target based on background subtraction
CN109255799B (en) Target tracking method and system based on spatial adaptive correlation filter
Dehghani et al. Small target detection and tracking based on the background elimination and Kalman filter
CN108305265B (en) Real-time processing method and system for weak and small target image
CN114373130A (en) Space-based infrared dark and weak moving target detection method
CN110390338B (en) SAR high-precision matching method based on nonlinear guided filtering and ratio gradient
CN110309808B (en) Self-adaptive smoke root node detection method in large-scale space
Mirabi et al. People tracking in outdoor environment using Kalman filter
CN108876807B (en) Real-time satellite-borne satellite image moving object detection tracking method
CN109308709B (en) Vibe moving target detection algorithm based on image segmentation
CN111881837B (en) Shadow extraction-based video SAR moving target detection method
CN110728695B (en) Video SAR moving target detection method based on image area accumulation
Zhang et al. Dehazing with improved heterogeneous atmosphere light estimation and a nonlinear color attenuation prior model
Juliu et al. Dim and small target detection based on improved spatio-temporal filtering

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