CN114255263B - Self-adaptive space dim and weak star identification method based on background identification - Google Patents
Self-adaptive space dim and weak star identification method based on background identification Download PDFInfo
- Publication number
- CN114255263B CN114255263B CN202111604791.5A CN202111604791A CN114255263B CN 114255263 B CN114255263 B CN 114255263B CN 202111604791 A CN202111604791 A CN 202111604791A CN 114255263 B CN114255263 B CN 114255263B
- Authority
- CN
- China
- Prior art keywords
- star
- background
- frame
- image
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 31
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 13
- 239000013598 vector Substances 0.000 claims description 11
- 238000003384 imaging method Methods 0.000 claims description 6
- 230000002776 aggregation Effects 0.000 claims description 2
- 238000004220 aggregation Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 claims description 2
- 230000003287 optical effect Effects 0.000 claims description 2
- 230000000007 visual effect Effects 0.000 abstract 1
- 238000001514 detection method Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 108091092878 Microsatellite Proteins 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
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
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- 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
Abstract
The invention discloses a background identification-based self-adaptive space dim star identification method, which comprises the following steps: (1) acquiring a plurality of continuous frames of star-sky images containing targets; (2) Carrying out background identification on each frame of star map, and extracting partial star information in a star field; (3) Respectively carrying out inter-frame association on a background fixed star and a suspected target; (4) Extracting the motion characteristics of a background fixed star and a suspected target in the star map, and adaptively calculating a target motion characteristic recognition threshold according to the background identification fixed star; the method uses the visual star images as the basis, uses the image recognition means to accurately extract star points, uses star recognition to identify star points in the background star, combines the inter-frame association relation of background star and suspected targets to extract motion characteristics and adaptively calculates related thresholds to realize the space dark and weak star recognition, improves the real-time performance and the robustness of the space dark and weak star recognition, and can better realize the on-orbit application.
Description
Technical Field
The invention relates to the technical field of weak and small target identification of complex starry sky backgrounds, in particular to a background identification-based self-adaptive space dim and weak star identification method.
Background
In recent years, with the emission of micro satellites such as constellations, star chains and the like in countries around the world, space dim stars are continuously growing. High-precision detection technology of space dim stars faces urgent demands. However, due to the hysteresis of foundation detection, the information such as the motion trail, azimuth and attitude of the space dark and weak star is difficult to track and calculate in real time. Meanwhile, the identification of the space dim star is a precondition for the calculation of the track, the azimuth, the gesture and the like.
At present, the method for identifying the space dark and weak stars mainly adopts a star removing mode to identify the space dark and weak stars, and cannot effectively utilize background star information. The star map matching-based remote space moving object recognition in the patent 201110095824.8 adopts a triangle star map recognition mode to recognize a star object, a characteristic star table is required to be constructed, and the object is mistakenly recognized as a background star with great probability. The recognition efficiency is low, and the threshold value needs to be set manually. The patent 201911043739.X also adopts a star removing mode to perform space dark and weak star identification, so that background star information cannot be effectively utilized, and a plurality of experience set thresholds are needed to complete target identification. The article "space dark and weak star detection and identification research of space dark and weak star" also adopts a star background inhibition and rejection mode to identify space dark and weak star, and neither relates to the space dark and weak star identification method based on background identification. In addition, the method of energy accumulation is adopted to identify the space dark and weak stars, multiple frames of images are required to accumulate energy, and the space dark and weak stars are identified according to the star point scribing. The method has the problems that false alarms are high or recognition rate is low when the motion speed is small or large, and the artificial experience threshold value is set. In summary, the existing space dim star identification method has the problems of dependence on priori information, experience threshold value, sensitivity to movement speed and the like.
Therefore, the method for identifying the space dark and weak star based on background identification is provided, does not depend on prior information, can adaptively adjust an identification threshold value, has the advantages of high identification rate, low false alarm rate and the like, and greatly improves the identification speed.
Disclosure of Invention
The invention solves the technical problems that: the space dim star identification method aiming at the space-based complex star sky background is provided. The method utilizes a background identification method to identify star information in a star map background, and simultaneously combines a reference star to construct motion feature vectors of each frame to realize association of star points of different categories among frames. Finally, a target motion characteristic recognition threshold is obtained through the self-adaptive motion characteristic threshold calculation method, and further recognition of the space dim star is achieved. Thereby improving the recognition rate and reducing the false alarm rate.
The technical scheme adopted by the invention is as follows: a background identification-based self-adaptive space dim star identification method comprises the following steps:
(1) Imaging the spatially dark and weak stars and acquiring a sequence of starry sky images I containing the target n . Spatially dark and weak stars are long-range weak and small objects that resemble the imaging characteristics of stars, using a two-dimensional gaussian function to describe the imaging characteristics of the stars and objects.
wherein u0 ，v 0 As the center position of the true star image, sigma PSF The gaussian radius represents the energy concentration of the point spread function, I (u, v) is the star map, u, v is the pixel coordinates of the star map, exp (·) is the exponential function based on e.
(2) And (5) identifying a star map background. Sequentially carrying out background recognition on the images obtained in the step (1) to extract the coordinate set { P } of all star point targets in the images k ＝(u k ,v k )}，u k 、v k Pixel coordinate values of the star point target in an image coordinate system; simultaneously, star point coordinate set { P } is extracted through star table, planet and moon orbit parameter pairs k Coarse classification is carried out to identify partial background sidereal target coordinate set as { S }, and m ＝(u m ,v m )}，u m 、v m pixel coordinate values in an image coordinate system for the star target; the suspected target set is { T ] n ＝(u n ,v n )},u n 、v n Pixel coordinate values of the suspected target in an image coordinate system;
according to the initial direction { alpha } 0 ，δ 0 ，φ 0 The right ascension and declination of the stars in the star list { alpha }, the right ascension and declination of the stars in the star list i ,δ i The coordinate position set { C } of the star on the image can be calculated i ＝(u i ,v i )}；
First, stars that can be imaged on an image are extracted according to the following formula: wherein FOV (field of view) x FOV for the field angle in the x-direction y Is the field angle in the y-direction.
Secondly, the coordinate position of the star on the image is calculated and extracted according to the following formula: alpha 0 ，δ 0 and φ0 Respectively the right ascension, the right ascension and the rotation angle of the current optical axis direction, alpha i and δi The right ascension and the right ascension of the star in the star table respectively, the width of the target surface of each image of W and H is high, and the resolution of each image is high, u i and vi The pixel coordinates of the stars on the image, respectively.
wherein ,xi To extract the non-rotated abscissa value of the star under the physical coordinate system of the image, y i Extracting fixed star as non-rotated ordinate value X in image physical coordinate system i To extract the abscissa pixel coordinate value of the star under the image coordinate system, Y i Extracting the ordinate pixel coordinate value of the star under the image coordinate system;
finally, the background fixed star in the image is identified according to the following formula, and the fixed star set { S } in the background n}： wherein Pk C for actually extracting star point coordinate set i To calculate the star coordinate set, id is the star, th d Is the star point distance threshold, d (P k ,C i ) For actually extracting the distance between the star point and calculating the star coordinates, id m The asterisks are the actual extracted star points.
Wherein: the positions of the planet and the moon in the solar system on the image are calculated according to the orbit parameters of the planet and the moon and the world to obtain the apparent ascension and the apparent ascension of the planet, so that the position coordinate set { M (M) of the planet and the moon on the image is calculated according to the formula j }. The optode warps and optode wefts of the planet and moon can be calculated according to astronomical algorithm.
The suspected target is composed of the set of residual star points after background recognition, i.e. { T } is obtained according to the following formula n }：
wherein ,d(Pk ,M j ) The distance between the planet and the moon coordinates is calculated for actually extracting the star points;
(3) Star point inter-frame correlation. For the I < th > obtained in the step (2) n Fixed star aggregation of frames
Suspected target set->
And I n+1 Sidereal set of frames->
Suspected target set->
And carrying out inter-frame association and determining the homonymy relation. The star set is associated according to star signs of stars, and the suspected target set is associated based on reference star feature vector similarity.
Firstly, the fixed star frame is related, fixed star sets of two frames are taken to find fixed stars with equal fixed star numbers as matched fixed stars, the fixed star matching relation is judged according to the following formula, and the fixed star coordinate set of the current frame and the previous frame is obtained to be the matched fixed star
wherein ,
for the sidereal set in frame n +.>
Is the fixed star set in the n+1th frame,>
star for star in the n frame star set,/-for star>
Star for the star in the star set for the n+1 frame.
Secondly, suspected target inter-frame association is carried out, and three associated stars are selected as reference stars according to brightness
and />
Respectively calculating three-dimensional feature vectors between suspected targets and reference stars in the previous frame and the current frame
and />
Star diagonal distance calculation formula:
wherein ,Si To select the reference star, T j As a candidate object, f is the focal length. According to the formula, three-dimensional feature vectors of candidate targets in different frames can be calculated respectively, similarity matching is carried out, and inter-frame association of suspected targets is completed if a similarity threshold is met.
Wherein Sim (A n ,A n+1 ) As a similarity function of the n-th frame suspected object and the n+1-th frame suspected object,
is the star diagonal distance between the kth suspected target and the reference star in the n+1th frame, +.>
And the star diagonal distance between the kth suspected target and the reference star in the nth frame.
(4) And (3) self-adaptive space dim star recognition, namely self-adaptively calculating a motion characteristic threshold according to the motion characteristics of the background recognition fixed star, so as to recognize the target. First, an initial image frame is selected from a sequence of images
Secondly, calculating the inter-frame association star and the suspected target obtained in the step (3) to obtain a motion feature set +.>
wherein 
and />
Representing the distance and direction of movement of the star point relative to the initial frame, respectively. Calculating the motion characteristics of all star points according to the above formula, and calculating the motion characteristics of the star based on the star obtained by the background identification in the step (2) and the motion characteristic calculation method in the step (3), wherein the calculated motion characteristics of the star are +.>
Adaptively calculating a motion characteristic threshold according to star background motion characteristics obtained by background identification:
wherein ThDL and ThDH The upper and lower boundaries of the motion distance threshold value obtained by adaptive calculation are respectively ThS L and ThSH The upper and lower boundaries of the motion direction threshold, respectively. Motion characteristic threshold value and motion characteristic of suspected target obtained by self-adaption calculation
The recognition result target is represented by the following formula:
wherein ,
for the target set +.>
Distance feature as a suspected object, +.>
Is the directional characteristic of the suspected object.
Compared with the prior art, the invention has the advantages that: the space dim star identification method aiming at the space-based complex star sky background is provided. The method utilizes a background identification method to project stars, planets, moon and the like in a star table on an image target surface according to the current gesture so as to identify background information. And respectively carrying out inter-frame association according to the identified star information and the suspected target information, wherein star association is adopted by the star, and the suspected target is associated based on the similarity of the reference star feature vectors. And finally, carrying out space dark and weak star recognition according to the star motion characteristic self-adaptive calculation motion characteristic recognition threshold value, and effectively utilizing star information in a star sky background, thereby improving the recognition rate of space dark and weak star recognition and reducing the false alarm rate.
Drawings
FIG. 1 is a specific flowchart of a background recognition-based adaptive spatial dim star recognition method;
FIG. 2 is a star point extraction of the star map of the present invention;
FIG. 3 is a star map background identification map of the present invention;
FIG. 4 is a diagram of the inter-frame correlation of the star map of the present invention;
FIG. 5 is a diagram showing the effect of object recognition according to the present invention.
Detailed Description
The process according to the invention is further illustrated below with reference to specific examples.
The invention discloses a background identification-based self-adaptive space dim star identification method, and a specific flow is shown in figure 1.
(1) Imaging a sky containing a target and acquiring a sequence image { I } n The acquired sequence image contains 4 spatially moving objects, the acquired sequence image { I } n N=10.
(2) Background identification, namely carrying out fixed star background feature identification on the image obtained in the step (1) to extract a background fixed star coordinate set { S } m Sum of the suspected target coordinates { T } and n specifically, the method comprises the following steps:
and (2-1) preprocessing the image to finish star point centroid extraction. And extracting star points of the star map according to the background gray level and connectivity of the star map. The background gray threshold is calculated from thb=n+3σ, where n is the average value of the image sensor random noise and σ is the root mean square value of the image sensor random noise. Then all star points are connected and extracted by judging 8 neighborhood, and the centroid position { P } of the star points is calculated according to centroid algorithm k }。
(2-2) extracting Star information of Star theory according to camera pose { α } 0 ，δ 0 ，φ 0 And the field angle phi 6 DEG of the cameraTo extract from the star list the right ascension and declination { alpha } of all stars in the field of view i ,δ i }。
(2-3) theoretical positions of stars, planets and moon on the image, and according to the sun right ascension and declination information { alpha } extracted in the step (2-2) i ,δ i The theoretical position of the star on the image target surface can be calculated by the angle phi 6 of the camera and the resolution 2048 multiplied by 2048.
(2-4) Star Point identification, theoretical position of the fixed star on the image according to the step (2-3) { (u) i ,v i ) And the star point { P } extracted in the step (2-1) k Identifying and obtaining a star set { S } in a star map background according to the following formula m }：
Wherein: the positions of the planets and the moon in the solar system on the image are calculated according to the orbit parameters of the planets and the moon and the world time to obtain the apparent warp and the apparent weft of the planets, so that the planets and the moon on the image are calculated according to the formulaPosition coordinate set { M on image j }. The optode warps and optode wefts of the planet and moon can be calculated according to astronomical calculation.
The suspected target is composed of the set of residual star points after background recognition, i.e. { T } is obtained according to the following formula n }：
(3) Inter-frame association, namely identifying the sequence images obtained in the step (1) according to the step (2) to respectively obtain a background fixed star coordinate set and a suspected target coordinate set in two continuous frame images; and respectively carrying out inter-frame association on a background fixed star set and a suspected target set in the two frames of images to obtain a homonymous relation matching pair, which is specifically as follows:
(3-1) associating between fixed frames, searching fixed stars with equal fixed stars and stars as matched fixed stars by using fixed stars set of two frames, judging fixed star matching relation according to the following formula, and obtaining fixed star coordinate set of matching current frame and previous frame as
(3-2) correlation between suspected target frames, selecting three correlated stars as reference stars according to brightness
and />
Respectively calculating three-dimensional feature vectors between suspected targets and reference stars in the previous frame and the current frame
and />
Star diagonal distance calculation formula:
wherein Si To select the reference star, T j As a candidate object, f is the focal length. According to the formula, three-dimensional feature vectors of candidate targets in different frames can be calculated respectively, similarity matching is carried out, and inter-frame association of suspected targets is completed if a similarity threshold is met.
(4) And (3) self-adaptive space dim star identification, namely acquiring the motion characteristics of the background star and the suspected target star according to the inter-frame association relation acquired in the step (3), and completing space motion target identification according to a self-adaptive characteristic threshold. The method comprises the following steps:
(4-1) initializing, namely selecting a first frame of initializing image frame from the sequence images, carrying out the background identification in the step (2) on the image frame, and obtaining an initial frame background fixed star coordinate set
And suspicious target coordinate set->
(4-2) self-adaptive motion characteristic threshold value, in the next frame of the initial image frame of the sequence image, performing step (2) background identification on the image to respectively obtain background fixed star coordinate sets of the image of the frame
And suspicious target set->
And calculating the motion characteristics of the sidereal associated with the initial frame and the current frame according to the homonymy obtained in the step (3)>
Thereby obtaining the motion characteristic threshold { ThD according to the following formula L ,ThD H ,ThS L ,ThS H }。
(4-3) identifying the space dim stars, calculating the motion characteristics of the associated suspected targets between the initial frame and the current frame
And identifying the space moving object by combining the self-adaptive motion characteristic threshold value.
(4-4) repeatedly performing the steps (4-2) and (4-3) on the sequence image according to the identification result until the space moving object is identified or the set maximum sequence frame count is reached, initializing the step (4-1) again, and performing the steps (4-2) and (4-3).
The invention, in part, is not described in detail in the manner known in the art. The foregoing is only illustrative of the present invention and is not to be construed as limiting thereof, but rather as various modifications, equivalent arrangements, improvements, etc., which fall within the spirit and principles of the present invention.
Claims (4)
1. A background identification-based self-adaptive space dim star identification method is characterized by comprising the following steps:
step (1): imaging the space dark and weak star, acquiring a star map containing the space dark and weak star, and continuously imaging to acquire a sequence star map I n ；
Step (2): performing star map background identification, namely sequentially performing background identification on the images obtained in the step (1) to extract coordinate sets { P } of all star point targets in the images k ＝(u k ,v k )}，u k 、v k For the pixel coordinate value of the star point target in the image coordinate system, simultaneously extracting a star point coordinate set { P ] through star table, planet and moon orbit parameter pairs k Coarse classification is carried out to identify partial background sidereal target coordinate set as { S }, and m ＝(u m ,v m )}，u m 、v m for the pixel coordinate value of the star target in the image coordinate system, the suspected target set is { T } n ＝(u n ,v n )}，u n 、v n Pixel coordinate values of the suspected target in an image coordinate system;
according to the initial direction { alpha } 0 ，δ 0 ，φ 0 The right ascension and declination of the stars in the star list { alpha }, the right ascension and declination of the stars in the star list i ,δ i The coordinate position set { C } of the star on the image can be calculated i ＝(u i ,v i )}：
First, stars that can be imaged on an image are extracted according to the following formula: wherein FOV (field of view) x FOV for the field angle in the x-direction y The angle of view in the y direction is:
secondly, the coordinate position of the star on the image is calculated and extracted according to the following formula: alpha 0 ，δ 0 and φ0 Respectively the right ascension, the right ascension and the rotation angle of the current optical axis direction, alpha i and δi The right ascension and the right ascension of the star in the star table respectively, the width of the target surface of each image of W and H is high, and the resolution of each image is high, u i and vi Pixel coordinates of stars on the image:
wherein ,xi To extract the non-rotated abscissa value of the star under the physical coordinate system of the image, y i Extracting fixed star as non-rotated ordinate value X in image physical coordinate system i To extract the abscissa pixel coordinate value of the star under the image coordinate system, Y i Extracting the ordinate pixel coordinate value of the star under the image coordinate system;
finally, the background fixed star in the image is identified according to the following formula, and the fixed star set { S } in the background m}： wherein Pk C for actually extracting star point coordinate set i To calculate the star coordinate set, id is the star, th d Is the star point distance threshold, d (P k ,C i ) For actually extracting the distance between the star point and calculating the star coordinates, id m Star for actually extracting star points:
wherein, the positions of the planet and the moon in the solar system on the image are calculated according to the orbit parameters of the planet and the moon and the world time to obtain the apparent ascension and the apparent ascension of the planet, so that the position coordinate set { M of the planet and the moon on the image is calculated according to the formula j The optode trails and the optode wefts of the planet and the moon can be calculated according to an astronomical algorithm;
the suspected target is composed of the set of residual star points after background recognition, i.e. { T } is obtained according to the following formula n }：
wherein ,d(Pk ,M j ) The distance between the planet and the moon coordinates is calculated for actually extracting the star points;
step (3): inter-star point frame correlation, for the I (th) acquired in the step (2) n Fixed star aggregation of frames
Suspected target set->
And I n+1 Sidereal set of frames->
Suspected target set->
Performing inter-frame association;
firstly, star frame-to-frame association is carried out, star sets of two frames are taken to find stars with equal star numbers as matched stars, and star matching relation is judged according to the following formulaObtaining the star coordinate set of the current frame and the previous frame as
wherein ,
for the sidereal set in frame n +.>
Is the fixed star set in the n+1th frame,>
star for star in the n frame star set,/-for star>
Star for star in the n+1st frame star set;
secondly, suspected target inter-frame association is carried out, and three associated stars are selected as reference stars according to brightness
and />
Calculating three-dimensional feature vectors between suspected targets and reference stars in the previous frame and the current frame respectively>
and />
Star diagonal distance calculation formula:
wherein ,Si To select the reference star, T j As candidate targets, f is a focal length, three-dimensional feature vectors of the candidate targets in different frames can be calculated respectively according to the formula, similarity matching is carried out, and inter-frame association of suspected targets is completed when a similarity threshold is met;
step (4): self-adaptive space dark and weak star recognition, namely self-adaptively calculating a motion characteristic threshold according to the motion characteristic of background recognition fixed star so as to recognize a target, and calculating the inter-frame association fixed star and the suspected target obtained in the step (3) to obtain a motion characteristic set
/>
wherein ,
and />
Separate tableDistance and direction of star point movement relative to the initial frame, < >>
and />
For the current frame star coordinates +.>
and />
Calculating the motion characteristics of all star points according to the above formula for the initial frame star point coordinates, and calculating the motion characteristics of the star based on the star obtained by the background identification in the step (2) and the motion characteristic calculation method in the step (3)>
Adaptively calculating a motion characteristic threshold according to star background motion characteristics obtained by background identification:
wherein ,ThDL and ThDH Respectively self-adaptingUpper and lower boundaries of the calculated movement distance threshold, thS L and ThSH The upper and lower boundaries of the motion direction threshold value are respectively, and the motion characteristic threshold value and the motion characteristic of the suspected target are obtained according to the self-adaptive calculation
The recognition result target is represented by the following formula:
wherein ,Ti n In order to set the objects in the set,
distance feature as a suspected object, +.>
Is the directional characteristic of the suspected object.
2. The method for identifying the self-adaptive space dim star based on background identification according to claim 1, wherein the method comprises the following steps: the starry sky background recognition equation S proposed in the step (2) m ＝{(u m ＝u k ，v m ＝v k ，id m ＝id)|d(P k ,C i )≤Th d Specifically, the star, the planet and the moon in the star table are calculated according to the current gesture, and the star background is recognized by combining the coordinate positions of the star, the planet and the moon on the image target surface and an error threshold value so as to achieve the effect of background recognition.
3. The method for identifying the self-adaptive space dim star based on background identification according to claim 1, wherein the method comprises the following steps: the star point inter-frame association method based on the reference star provided in the step (3) constructs three-dimensional feature vectors between suspected targets and the reference star in different image frames
and />
And judging the similarity of the inter-frame feature vectors to realize inter-frame star point association.
4. The method for identifying the self-adaptive space dim star based on background identification according to claim 1, wherein the method comprises the following steps: and (3) the self-adaptive motion characteristic threshold value calculation method provided in the step (4) is used for obtaining sidereal motion characteristic information according to background identification and self-adaptively calculating the motion characteristic threshold value which the background sidereal should have, so that the target is identified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111604791.5A CN114255263B (en) | 2021-12-24 | 2021-12-24 | Self-adaptive space dim and weak star identification method based on background identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111604791.5A CN114255263B (en) | 2021-12-24 | 2021-12-24 | Self-adaptive space dim and weak star identification method based on background identification |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114255263A CN114255263A (en) | 2022-03-29 |
| CN114255263B true CN114255263B (en) | 2023-05-26 |
Family
ID=80797750
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111604791.5A Active CN114255263B (en) | 2021-12-24 | 2021-12-24 | Self-adaptive space dim and weak star identification method based on background identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114255263B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114926351B (en) * | 2022-04-12 | 2023-06-23 | 荣耀终端有限公司 | Image processing method, electronic device, and computer storage medium |
| CN116755169B (en) * | 2023-06-13 | 2024-04-30 | 南京航空航天大学 | Small target detection method and system based on star map identification and brightness priori information |
| CN116933567B (en) * | 2023-09-15 | 2024-02-02 | 中国科学院光电技术研究所 | Space-based complex multi-scene space target simulation data set construction method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109708648A (en) * | 2018-11-27 | 2019-05-03 | 上海航天控制技术研究所 | A kind of classification discrimination method of spatial movement point target |
| CN111126131A (en) * | 2019-10-30 | 2020-05-08 | 北京控制工程研究所 | High-efficiency dark and weak space target identification method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190385282A1 (en) * | 2018-06-18 | 2019-12-19 | Drvision Technologies Llc | Robust methods for deep image transformation, integration and prediction |
-
2021
- 2021-12-24 CN CN202111604791.5A patent/CN114255263B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109708648A (en) * | 2018-11-27 | 2019-05-03 | 上海航天控制技术研究所 | A kind of classification discrimination method of spatial movement point target |
| CN111126131A (en) * | 2019-10-30 | 2020-05-08 | 北京控制工程研究所 | High-efficiency dark and weak space target identification method |
Non-Patent Citations (3)
| Title |
|---|
| Yangting Lin 等.Concepts of the Small Body Sample Return Missions - the 1st 10 Million Year Evolution of the Solar System.Space Science Reviews.2020,全文. * |
| 张文明 等."小型星模拟器中星图动态显示系统的设计 ".光电工程.2000,全文. * |
| 樊萌."基于Scilab/Xcos的星敏感器星图识别仿真研究 ".中国优秀硕士学位论文全文数据库 (工程科技Ⅱ辑).2020,C031-54. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114255263A (en) | 2022-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114255263B (en) | Self-adaptive space dim and weak star identification method based on background identification | |
| CN111626290B (en) | Infrared ship target detection and identification method under complex sea surface environment | |
| KR100780464B1 (en) | Adaptive star centroiding method in star sensor based | |
| CN106651904B (en) | A kind of more extraterrestrial target method for capturing and tracing of width size range | |
| CN111506759B (en) | Image matching method and device based on depth features | |
| Xi et al. | Space debris detection using feature learning of candidate regions in optical image sequences | |
| CN107504966B (en) | Method for extracting navigation star points in daytime cloud environment | |
| CN102254147B (en) | Method for identifying long-distance space motion target based on stellar map matching | |
| CN113888461A (en) | Method, system and equipment for detecting defects of hardware parts based on deep learning | |
| CN111275740B (en) | Satellite video target tracking method based on high-resolution twin network | |
| Liang et al. | Robust sea-sky-line detection for complex sea background | |
| CN116486243A (en) | DP-ViT-based sonar image target detection method | |
| Wang et al. | Robust detection of infrared maritime targets for autonomous navigation | |
| CN111062310B (en) | Few-sample unmanned aerial vehicle image identification method based on virtual sample generation | |
| Xiao et al. | Multiclass object detection in UAV images based on rotation region network | |
| CN108519083A (en) | A kind of capture of space non-cooperative multiple target and track algorithm | |
| CN108320310A (en) | Extraterrestrial target 3 d pose method of estimation based on image sequence | |
| CN115063428B (en) | Spatial dim small target detection method based on deep reinforcement learning | |
| CN114332444B (en) | Complex star sky background target identification method based on incremental drift clustering | |
| CN113532446B (en) | Star sensor stray light resistant star point extraction method and device based on iterative traversal | |
| Kumar et al. | Ship detection based on faster R-CNN in SAR imagery by anchor box optimization | |
| CN115760984A (en) | Non-cooperative target pose measurement method based on monocular vision by cubic star | |
| CN111126131B (en) | High-efficiency dark and weak space target identification method | |
| Dong | Faint moving small target detection based on optical flow method | |
| Peerapong et al. | Space debris tracking strategy through Monte-Carlo based track-before-detect framework |
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 |