CN110793529A - Quick matching star map identification method - Google Patents
Quick matching star map identification method Download PDFInfo
- Publication number
- CN110793529A CN110793529A CN201911307702.3A CN201911307702A CN110793529A CN 110793529 A CN110793529 A CN 110793529A CN 201911307702 A CN201911307702 A CN 201911307702A CN 110793529 A CN110793529 A CN 110793529A
- Authority
- CN
- China
- Prior art keywords
- star
- coordinate system
- distance
- stars
- fixed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/24—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for cosmonautical navigation
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- Astronomy & Astrophysics (AREA)
- Automation & Control Theory (AREA)
- General Physics & Mathematics (AREA)
- Navigation (AREA)
Abstract
The invention relates to a quick matching star map identification method. The method of the invention extracts the characteristics of the known simple star catalogue, constructs a brand new navigation star catalogue, and realizes matching calculation and identification of fixed stars in the field of view of the star sensor by using the star map identification algorithm of the invention. In the identification process, the correlation analysis of the fixed star-star pair is added, so that the identification accuracy is improved; the defects of large storage space requirement, low reaction speed, poor instantaneity and the like of the conventional identification algorithm are overcome.
Description
Technical Field
The invention is applied to the technical field of star map identification in star sensors, and particularly relates to a quick matching star map identification method.
Background
The star sensor is the most widely used celestial body sensor in the current astronomical navigation system. The method takes the fixed star as a reference source and provides accurate real-time data for the control and astronomical navigation of the spacecraft. In the star sensor, the all-day adaptive star map identification is also a very critical technology. The star map identification is to acquire a star map plane in a view field through a star sensor, and correspondingly match the star map with a reference star in a navigation star table to finish the identification of a fixed star in the view field. The complete star atlas identification system generally comprises navigation star atlas construction, image acquisition and preprocessing, algorithm matching identification and the like.
The existing star map recognition mainly comprises a triangle positioning method, a neural network matching algorithm and the like. However, the existing recognition algorithm ignores the practical application performance in the process of continuously improving the accuracy. The triangle algorithm has the characteristics of simplicity and easiness in implementation, but because the fixed stars are large in number, the triangle navigation star table formed on the basis of the fixed stars needs to be stored in a large database, and the fact that hardware equipment is actually applied and a memory is externally connected is not a good choice. Along with the rapid development of neural network algorithms, neural network matching is brought forward. However, the neural network matching algorithm is based on simulation experiments at present, and because of numerous types of fixed stars, the whole network needs a large number of neuron connections, the training time is long, meanwhile, forward calculation also needs a large number of calculation resources and storage resources, and the method cannot be effectively used in the actual application of the star sensor.
Disclosure of Invention
The invention aims to provide a quick matching star atlas identification method, which is characterized by extracting the characteristics of a known simple star atlas, constructing a brand new navigation star atlas, realizing matching calculation by using the star atlas identification algorithm of the invention and identifying fixed stars in the field of view of a star sensor; in the identification process, the correlation analysis of the fixed star-star pair is added, so that the identification accuracy is improved; the defects of large storage space requirement, low reaction speed, poor instantaneity and the like of the conventional identification algorithm are overcome.
In order to achieve the purpose, the technical scheme of the invention is as follows: a quick matching star map identification method comprises the following steps:
step S1, constructing a navigation star chart:
firstly, screening the grades of fixed stars and stars in an original star catalogue, and eliminating fixed stars with low star levels to avoid deviation in star map identification calculation;
secondly, let O1-X1Y1Z1Representing a star sensor coordinate system, and O-XYZ representing a celestial coordinate system; o is the center of the earth, O1Is the center of the star sensor optical system; according to the definition of a celestial coordinate system, an origin O is coincident with the geocenter, the direction of an OZ axis points to the north pole from the geocenter, the direction of an OX axis points to the spring equinox from the geocenter, and the direction of an OY axis can be determined according to a right-hand coordinate system; O-XYZ coordinate and O1-X1Y1Z1The relationship between the coordinates can be expressed as:
[X1Y1Z1]＝R[X Y Z]
according to the rotation sequence of the coordinates, the rotation is completed to the coordinate system O-XYZ of the celestial coordinate system to the coordinate system O of the image plane1-X1Y1Z1The transformation of (1); at this time, the OY axis and OY1The axes are corresponding to each other, and the OZ axis is corresponding to the OZ1Axis corresponding to OX axis1The axes are corresponding;
let the right ascension and declination of the observation star be (α)i,δi) And according to the conversion relation between the spherical coordinate system and the rectangular coordinate system, the unit vector of the fixed star under the rectangular coordinate system of the celestial sphere is obtained as follows:
[X Y Z]＝[cosδicosαicosδisinαisinδi]
the coordinate positions of the observation stars in the space rectangular coordinate system can be obtained by arranging the two formulas;
performing distance pre-calculation on the solved coordinate values of the fixed stars to obtain the distance between each star pair; taking the star pair distance as a key object, constructing a brand new navigation star list, wherein each sample also comprises corresponding serial numbers of two fixed stars, and deleting the rest information;
step S2, star map information acquisition:
firstly, preprocessing an image of a star atlas;
secondly, acquiring information of the image, and constructing a star sensor mirror image plane coordinate system by taking the pixel size as a basic unit;
finally, since the star map is imaged in a planar coordinate system, it is necessary to determine the pixel size d included in the star sensorh,dvThe information of the focal length f is subjected to coordinate transformation once through the relation of projection transformation to obtain a coordinate value of the fixed star before projection under a rectangular coordinate system;
step S3, star map identification:
firstly, preprocessing star map data collected by a star sensor, and acquiring data of a collected fixed star;
secondly, based on a celestial sphere rectangular coordinate system, a distance formula is reused
Calculating the distance of each star pair to construct a star pair distance table corresponding to the star map; wherein, Xi、XjRespectively the coordinate values of the fixed stars i and j in the X dimension under the rectangular coordinate system of the celestial sphere, and Yi、YjAre coordinate values of fixed stars i and j in Y dimension under a rectangular coordinate system of celestial sphere, Zi、ZjCoordinate values of fixed stars i and j in the Z dimension under the celestial sphere rectangular coordinate system respectively;
then, the first K star pairs which are most similar to the star list to be identified are searched from the navigation star table to serve as the adjacent star pairs of the identified constant star pair, and a variance formula is utilized
Wherein the sample distance features of the real-time star pair are expressed as x ═ { x ═ x1,x2,x3,x4,...,xnThe distance sample characteristics of the satellite pairs of the matched navigation star table areCalculating the variance between the component distance value of each real-time satellite pair and the distance in the navigation star table, and comparing the component distance value with the distance in the navigation star table to realize distance variance matching;
finally, the obtained star-to-pair distance variance value set is processedPriority sorting, using priority queue to select the minimum variance front N item set delta ═ delta { (delta)1,δ2,δ3,...,δN}；
Step S4, identification correlation analysis:
according to the selected minimum variance front N item set delta ═ delta { (delta)1,δ2,δ3,...,δN}; matching the star number (m) common to each star pair1,m2,m3,..) to search neighboring stars around the reference star by using these stars as the reference stars.
Compared with the prior art, the invention has the following beneficial effects: the method has the advantages of small storage space requirement, high identification speed, high accuracy and convenience for carrying on a hardware system; on the premise of ensuring the accuracy, the star map identification system can provide real-time all-weather star map identification for the star sensor, and the main advantages are summarized as follows:
storage space requirement aspect: aiming at the problems of large information amount and high storage space requirement of the navigation star catalogue, the method carries out feature extraction on the traditional star catalogue, eliminates star information irrelevant to the algorithm and excessively low stars such as stars and the like, and constructs a brand new navigation star catalogue by taking the star pair distance as a sample key object, thereby reducing the memory requirement of a navigation star catalogue database;
star map recognition algorithm aspect: aiming at the problems of high star map identification rate and high speed, the method measures the star similarity based on the integration of distance variance matching and nearest neighbor correlation analysis, carries out method design optimization aiming at the actual situation, identifies the star number in the field of view through the correlation matching analysis of the star pair in the constructed navigation star table, and ensures the accuracy of the final identification result.
Drawings
Fig. 1 is a flow chart of star map recognition.
Fig. 2 is a planar imaging coordinate diagram of the star sensor.
FIG. 3 is a block diagram of an implementation of a star map identification algorithm.
FIG. 4 is a flow chart of a star map identification algorithm.
Detailed Description
The technical scheme of the invention is specifically explained below with reference to the accompanying drawings.
The invention provides a quick matching star map identification method, which comprises the following steps:
step S1, constructing a navigation star chart:
firstly, screening the grades of fixed stars and stars in an original star catalogue, removing fixed stars with low star levels and the like, and avoiding deviation in star map identification calculation, wherein a specific star threshold is determined by satellite sensor performance parameters;
secondly, let O1-X1Y1Z1Representing a star sensor coordinate system, and O-XYZ representing a celestial coordinate system; o is the center of the earth, O1Is the center of the star sensor optical system; according to the definition of a celestial coordinate system, an origin O is coincident with the geocenter, the direction of an OZ axis points to the north pole from the geocenter, the direction of an OX axis points to the spring equinox from the geocenter, and the direction of an OY axis can be determined according to a right-hand coordinate system; O-XYZ coordinate and O1-X1Y1Z1The relationship between the coordinates can be expressed as:
[X1Y1Z1]＝R[X Y Z]
according to the rotation sequence of the coordinates, the rotation is completed to the coordinate system O-XYZ of the celestial coordinate system to the coordinate system O of the image plane1-X1Y1Z1The transformation of (1); at this time, the OY axis and OY1The axes are corresponding to each other, and the OZ axis is corresponding to the OZ1Axis corresponding to OX axis1The axes are corresponding;
let the right ascension and declination of the observation star be (α)i,δi) And according to the conversion relation between the spherical coordinate system and the rectangular coordinate system, the unit vector of the fixed star under the rectangular coordinate system of the celestial sphere is obtained as follows:
[X Y Z]＝[cosδicosαicosδisinαisinδi]
the coordinate positions of the observation stars in the space rectangular coordinate system can be obtained by arranging the two formulas;
performing distance pre-calculation on the solved coordinate values of the fixed stars to obtain the distance between each star pair; taking the star pair distance as a key object, constructing a brand new navigation star table, wherein each sample comprises corresponding serial numbers of two fixed stars, and deleting the rest information;
step S2, star map information acquisition:
firstly, preprocessing an image of a star atlas;
secondly, acquiring information of the image, and constructing a star sensor mirror image plane coordinate system by taking the pixel size as a basic unit;
finally, since the star map is imaged in a planar coordinate system, it is necessary to determine the pixel size d included in the star sensorh,dvThe information of the focal length f is subjected to coordinate transformation once through the relation of projection transformation to obtain a coordinate value of the fixed star before projection under a rectangular coordinate system;
step S3, star map identification:
firstly, preprocessing star map data collected by a star sensor, and acquiring data of a collected fixed star;
secondly, based on a celestial sphere rectangular coordinate system, a distance formula is reused
Calculating the distance of each star pair to construct a star pair distance table corresponding to the star map; wherein, Xi、XjRespectively the coordinate values of the fixed stars i and j in the X dimension under the rectangular coordinate system of the celestial sphere, and Yi、YjAre coordinate values of fixed stars i and j in Y dimension under a rectangular coordinate system of celestial sphere, Zi、ZjCoordinate values of fixed stars i and j in the Z dimension under the celestial sphere rectangular coordinate system respectively;
then, the first K star pairs which are most similar to the star list to be identified are searched from the navigation star table to serve as the adjacent star pairs of the identified constant star pair, and a variance formula is utilized
Sample distance feature table of real-time star pairsIs shown as x ═ x1,x2,x3,x4,...,xnThe distance sample characteristics of the satellite pairs of the matched navigation star table are
Calculating the variance between the component distance value of each real-time satellite pair and the distance in the navigation star table, and comparing the component distance value with the distance in the navigation star table to realize distance variance matching;
and finally, carrying out priority ordering on the obtained star pair distance variance value set, and selecting the N item set delta before the minimum variance as { delta } by using a priority queue1,δ2,δ3,...,δN}；
Step S4, identification correlation analysis:
according to the selected minimum variance front N item set delta ═ delta { (delta)1,δ2,δ3,...,δN}; matching the star number (m) common to each star pair1,m2,m3,..) to search neighboring stars around the reference star by using these stars as the reference stars.
The following is a specific implementation of the present invention.
The star atlas identification flow chart of the invention is shown in figure 1. The star map identification system mainly comprises four parts of construction of a navigation star map, star map acquisition, star map identification and correlation analysis, and the specific functions of each part are as follows:
1. constructing a navigation star table:
first, since the field of view of the star sensor is limited by the illumination light source, all stars cannot be captured sufficiently, and errors are caused by weak stars such as stars being partially distant from the stars. Therefore, the grades of stars, such as stars and stars, in the original star catalogue are screened, the stars with too low stars, such as stars and the like, are removed, and deviation caused in the calculation of star map identification is avoided.
Secondly, the position information of the fixed star in the star table is expressed by the right ascension and the declination in the celestial coordinate system, and the star sensor images the star point under the star sensor coordinate system, so that the right ascension and the declination of the observed star need to be converted into coordinate values. Let O be1-X1Y1Z1Representing the star sensor coordinate system, O1-X1Y1Z1Representing a celestial coordinate system. O is the center of the earth, O1Which is the center of the star sensor optical system. The required result is obtained by two times of coordinate axis transformation, and because the distance between the fixed star and the earth is very far, the error caused by coordinate translation is negligible, and only the rotation transformation between coordinate systems is considered.
According to the definition of the celestial coordinate system, the origin O is coincident with the geocenter, the direction of the OZ axis is directed to the north pole from the geocenter, the direction of the OX axis is directed to the spring equinox from the geocenter, and the direction of the OY axis can be determined according to a right-hand coordinate system. O-XYZ coordinate and O1-X1Y1Z1The relationship between the coordinates can be expressed as:
[X1Y1Z1]＝R[X Y Z]
according to the rotation sequence of the coordinates, the rotation is completed to the coordinate system O-XYZ of the celestial coordinate system to the coordinate system O of the image plane1-X1Y1Z1And (4) transforming. At this time, the OY axis and OY1The axes are corresponding to each other, and the OZ axis is corresponding to the OZ1Axis corresponding to OX axis1The axes are corresponding;
let the right ascension and declination of the observation star be (α)i,δi) And according to the conversion relation between the spherical coordinate system and the rectangular coordinate system, the unit vector of the fixed star under the rectangular coordinate system of the celestial sphere is obtained as follows:
[X Y Z]＝[cosδicosαicosδisinαisinδi]
the coordinate positions of the observation stars in the space rectangular coordinate system can be obtained by arranging the two formulas;
and performing distance pre-calculation on the solved star coordinate values to obtain the distance between each star pair. And constructing a brand new navigation star table by taking the star pair distance as a key object, wherein each sample comprises corresponding serial numbers of two stars. At the same time, the remaining information is no longer retained in order to reduce the storage space requirements. The sample structure content of the navigation star catalogue data is shown in table 1.
TABLE 1
2. Acquisition of star map information:
in the process of collecting images by the star sensor, noise is introduced in the image generation due to the influence of signals, working environment, circuit structure and the like. In order to eliminate unnecessary and redundant interference information in star map imaging, image preprocessing is carried out on star maps before star map identification is carried out, and relevant key information is obtained. Secondly, the image sensor has limited brightness perception capability on stars, and some stars with weak brightness are unstable in imaging in the star sensor, which causes certain errors in subsequent star map identification. Thus, because of the presence of these interfering stars within the field of view of the star sensor, such erratic factors as interfering stars are also eliminated.
Since the star map is mainly gaussian noise, a gaussian filter can be used for noise elimination. And scanning each pixel point of the star map image by using a Gaussian template so as to determine the weighted average gray value of the pixels in the neighborhood. By two-dimensional Gaussian function
Where δ is the standard deviation. And discretizing the two-dimensional Gaussian function to obtain a numerical value serving as a template coefficient to obtain the Gaussian template. The template middle value is the central element, and the pixel value will change after calculation. The image processed by the Gaussian smoothing filter can smooth the imaging picture and reduce noise. Secondly, in order to solve errors caused by interference stars and screen out excessively low stars such as stars, the limit of star sensitivity of the star sensor is assumed to be 5.5Mv, but space radiation has practical influence on a receiving optical system and final performance evaluation of the star sensor, and a star threshold value can be set to be 5 Mv. Then, information acquisition can be carried out on the image, and a star sensor mirror image plane coordinate system is constructed by taking the pixel size as a basic unit. Finally, due to the starThe image is imaged in a plane coordinate system, so that the image needs to be measured according to the pixel size d in the star sensorh,dvAnd the focal length f and other information are subjected to coordinate transformation once through the relation of projection transformation to obtain coordinate values of the fixed star before projection under the rectangular coordinate system. Fig. 2 is a planar imaging coordinate diagram of the star sensor.
3. Star map recognition algorithm:
the first layer of the present star map identification algorithm will use a distance variance matching algorithm. The distance variance matching algorithm is a simple and easy classification algorithm without parameter estimation and training, is a relatively mature method in theory, and is one of the simplest machine learning algorithms. The star pairs in the star map are used as matching preset values, and the star pairs in the navigation star list are subjected to matching calculation with the preset values to obtain the nearest K groups of star pair numbers. The method is simple and easy to implement, and can have certain robustness through selection of K. And because the navigation star catalogue is reconstructed and the star map is preprocessed, a certain amount of calculation and resource storage are reduced. Experimental results show that the improved star map recognition algorithm used in the star sensor has good pre-recognition effect.
The improved star map recognition algorithm can not only ensure the recognition accuracy, but also ensure the execution efficiency of algorithm calculation. The star atlas identification algorithm computation implementation block diagram of the invention is shown in fig. 3. The implementation of the algorithm has the following 4 phases: firstly, preprocessing star map data acquired in a star sensitive period, and acquiring data of acquired stars; secondly, based on a celestial sphere rectangular coordinate system, a distance formula is reusedAnd (4) carrying out distance calculation on each star pair to construct a star pair distance table corresponding to the star map. Wherein, Xi、XjRespectively the coordinate values of the fixed stars i and j in the X dimension under the rectangular coordinate system of the celestial sphere, and Yi、YjRespectively are coordinate values of fixed stars i and j on X dimension under a celestial sphere rectangular coordinate system, Zi、ZjCoordinate values of fixed stars i and j in the X dimension under a celestial sphere rectangular coordinate system respectively; then look up from the navigation star tableFinding the first K fixed star pairs most similar to the fixed star list to be identified as the adjacent star pairs of the identified fixed star pair, and utilizing a variance formula
Wherein the sample distance features of the real-time star pair are expressed as x ═ { x ═ x1,x2,x3,x4,...,xnThe distance sample characteristics of the satellite pairs of the matched navigation star table are
Calculating the variance between the component distance value of each real-time satellite pair and the distance in the navigation star table, and comparing the component distance value with the distance in the navigation star table to realize distance variance matching; and finally, obtaining a star pair distance variance value set according to the calculation result, carrying out priority ordering on the star pair distance variance value set, and selecting the N item set delta before the minimum variance as the { delta } by using a priority queue1,δ2,δ3,...,δNAnd providing basic data for subsequent identification correlation analysis.
4. And (3) identifying correlation analysis:
after all fixed stars in the field of view of the star sensor are subjected to star map identification calculation, due to the fact that the field of view is fixed, the positions of all fixed stars have certain relevance, and the second-layer nearest reanalysis can be performed by means of the relevance among all fixed stars. Meanwhile, in the same field of view, the optimal star number results judged by all star pairs based on the Euclidean distance formula may overlap. Reanalysis of the results can also avoid errors.
The distance variance matching is used as a first-layer coarse matching of the feature matching, and the nearest neighbor correlation analysis completes the identification matching of a second layer by combining the actual situation and the previous first-layer coarse matching, namely the nearest neighbor correlation analysis performs re-identification on the basis of the distance analysis variance matching. And calculating by an Euclidean distance formula to obtain a variance result, and extracting N groups of data with minimum variance of each group of star pairs. Matching star numbers (m) common to all star pairs while determining the optimal result of matching by priority1,m2,m3,...). Using these stars as referenceAnd the fixed star is used for searching adjacent stars around the reference star. Multiple tests show that the double-layer identification analysis is utilized to form a complementary identification mode, so that the identification performance is improved, and the identification system has higher reliability. Therefore, the method adopts distance variance matching and nearest neighbor analysis integration, and further improves the accuracy and reliability of star atlas identification by carrying out method design optimization aiming at actual conditions.
The specific implementation example of the invention is as follows:
removing lower stars such as stars
The right star is arranged at right ascension and declination under a celestial coordinate system (α)i,δi) Performing coordinate transformation
Using Euclidean distance formula
Calculating the distance between each constant star
And constructing a navigation star table with the star pair as a key value and the star pair distance as a value, wherein each sample also comprises a star corresponding number.
Star map information acquisition process
Setting the size t of a filtering window template and the mean square error delta of a Gaussian distribution function to obtain a Gaussian filter G, and eliminating image noise by using the obtained Gaussian filter G
Screening out sidereal stars with imaging brightness lower than a certain threshold value
In pixel size dh,dvAs a basic unit, constructing a star sensor mirror image plane coordinate system
Obtaining the space coordinate value (x) of each star through coordinate transformation by using the relation of projection transformationi,yi,zi)
The star map identification algorithm flow chart is shown in fig. 4, and the work flow is as follows:
obtaining coordinate value (x) of each stari,yi,zi) Building a star map and star pair list
Using Euclidean distance formulaCalculating each of the pair distances in the current pair list
Traversing the star map and star pair list and utilizing the variance formula
Variance matching calculation is carried out on each star pair distance in star map imaging and the star pair distances in the known navigation star table
Carrying out priority arrangement on the variances of each group of star pairs, and acquiring a variance set delta of the minimum front N item, namely [ delta ]1,δ2,δ3,...,δN}
And performing correlation analysis and outputting a final star number.
Table 2 shows the star map recognition results of the present invention.
TABLE 2
The above are preferred embodiments of the present invention, and all changes made according to the technical scheme of the present invention that produce functional effects do not exceed the scope of the technical scheme of the present invention belong to the protection scope of the present invention.
Claims (1)
1. A quick matching star map identification method is characterized by comprising the following steps:
step S1, constructing a navigation star chart:
firstly, screening the grades of fixed stars and stars in an original star catalogue, and eliminating fixed stars with low star levels to avoid deviation in star map identification calculation;
secondly, let O1-X1Y1Z1Representing a star sensor coordinate system, and O-XYZ representing a celestial coordinate system; o is the center of the earth, O1Is the center of the star sensor optical system; according to the definition of a celestial coordinate system, an origin O is coincident with the geocenter, the direction of an OZ axis points to the north pole from the geocenter, the direction of an OX axis points to the spring equinox from the geocenter, and the direction of an OY axis can be determined according to a right-hand coordinate system; O-XYZ coordinate and O1-X1Y1Z1The relationship between the coordinates can be expressed as:
[X1Y1Z1]＝R[X Y Z]
according to the rotation sequence of the coordinates, the rotation is completed to the coordinate system O-XYZ of the celestial coordinate system to the coordinate system O of the image plane1-X1Y1Z1The transformation of (1); at this time, the OY axis and OY1The axes are corresponding to each other, and the OZ axis is corresponding to the OZ1Axis corresponding to OX axis1The axes are corresponding;
let the right ascension and declination of the observation star be (α)i,δi) And according to the conversion relation between the spherical coordinate system and the rectangular coordinate system, the unit vector of the fixed star under the rectangular coordinate system of the celestial sphere is obtained as follows:
[X Y Z]＝[cosδicosαicosδisinαisinδi]
the coordinate positions of the observation stars in the space rectangular coordinate system can be obtained by arranging the two formulas;
performing distance pre-calculation on the solved coordinate values of the fixed stars to obtain the distance between each star pair; taking the star pair distance as a key object, constructing a brand new navigation star list, wherein each sample also comprises corresponding serial numbers of two fixed stars, and deleting the rest information;
step S2, star map information acquisition:
firstly, preprocessing an image of a star atlas;
secondly, acquiring information of the image, and constructing a star sensor mirror image plane coordinate system by taking the pixel size as a basic unit;
finally, since the star map is imaged in a planar coordinate system, it is necessary to determine the pixel size d included in the star sensorh,dvThe information of the focal length f is subjected to coordinate transformation once through the relation of projection transformation to obtain a coordinate value of the fixed star before projection under a rectangular coordinate system;
step S3, star map identification:
firstly, preprocessing star map data collected by a star sensor, and acquiring data of a collected fixed star;
secondly, based on a celestial sphere rectangular coordinate system, a distance formula is reused
Calculating the distance of each star pair to construct a star pair distance table corresponding to the star map; wherein, Xi、XjRespectively the coordinate values of the fixed stars i and j in the X dimension under the rectangular coordinate system of the celestial sphere, and Yi、YjAre coordinate values of fixed stars i and j in Y dimension under a rectangular coordinate system of celestial sphere, Zi、ZjCoordinate values of fixed stars i and j in the Z dimension under the celestial sphere rectangular coordinate system respectively;
and then searching the first K star pairs which are most similar to the star list to be identified from the navigation star table as the adjacent star pairs of the identified constant star pair, wherein the sample distance characteristic of the real-time star pair is expressed as x ═ { x ═ x1,x2,x3,x4,...,xnThe distance sample characteristics of the satellite pairs of the matched navigation star table are
Using the formula of variance
Calculating the variance between the distance value of each real-time satellite pair and the distance in the navigation star table, and comparing the distance value with the distance in the navigation star table to realize distance variance matching;
and finally, carrying out priority ordering on the obtained star pair distance variance value set, and selecting the N item set delta before the minimum variance as { delta } by using a priority queue1,δ2,δ3,...,δN}；
Step S4, identification correlation analysis:
according to the selected minimum variance front N item set delta ═ delta { (delta)1,δ2,δ3,...,δN}; matching the star number (m) common to each star pair1,m2,m3,..), mixing the above-mentioned materialsAnd taking the fixed star as a reference fixed star, and searching adjacent stars around the reference star according to the fixed star.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911307702.3A CN110793529B (en) | 2019-12-18 | 2019-12-18 | Quick matching star map identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911307702.3A CN110793529B (en) | 2019-12-18 | 2019-12-18 | Quick matching star map identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110793529A true CN110793529A (en) | 2020-02-14 |
| CN110793529B CN110793529B (en) | 2022-06-03 |
Family
ID=69448533
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911307702.3A Active CN110793529B (en) | 2019-12-18 | 2019-12-18 | Quick matching star map identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110793529B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112529809A (en) * | 2020-12-15 | 2021-03-19 | 航天信息股份有限公司 | Starry sky image color difference correction method and device, storage medium and aerospace observation equipment |
| CN112665579A (en) * | 2020-12-01 | 2021-04-16 | 中国人民解放军国防科技大学 | Star map identification method and device based on geometric verification |
| CN114252088A (en) * | 2020-09-25 | 2022-03-29 | 南京理工大学 | Star map matching semi-physical verification method based on collimator |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102072730A (en) * | 2010-11-04 | 2011-05-25 | 北京航空航天大学 | Method for identifying triangular star map |
| CN102128623A (en) * | 2010-12-02 | 2011-07-20 | 哈尔滨工业大学 | Fast star map identification method of star sensor |
| US20120274810A1 (en) * | 2011-04-28 | 2012-11-01 | Honeywell International Inc. | Inverse star tracker using psuedo-random overlapping circles |
| CN103453905A (en) * | 2013-08-07 | 2013-12-18 | 北京航空航天大学 | Star pattern recognition method based on principal component analysis of plane triangles |
| CN105371870A (en) * | 2015-11-27 | 2016-03-02 | 北京空间飞行器总体设计部 | Star map data based method for measurement of in-orbit precision of star sensor |
| CN109238269A (en) * | 2018-09-05 | 2019-01-18 | 南京航空航天大学 | A kind of star pattern matching method across the fusion of discontinuous visual field |
| CN109359577A (en) * | 2018-10-08 | 2019-02-19 | 福州大学 | A kind of Complex Background number detection system based on machine learning |
| CN109813303A (en) * | 2019-03-08 | 2019-05-28 | 北京航空航天大学 | A kind of method for recognising star map for not depending on calibrating parameters based on the ballot of angle mould formula cluster |
-
2019
- 2019-12-18 CN CN201911307702.3A patent/CN110793529B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102072730A (en) * | 2010-11-04 | 2011-05-25 | 北京航空航天大学 | Method for identifying triangular star map |
| CN102128623A (en) * | 2010-12-02 | 2011-07-20 | 哈尔滨工业大学 | Fast star map identification method of star sensor |
| US20120274810A1 (en) * | 2011-04-28 | 2012-11-01 | Honeywell International Inc. | Inverse star tracker using psuedo-random overlapping circles |
| CN103453905A (en) * | 2013-08-07 | 2013-12-18 | 北京航空航天大学 | Star pattern recognition method based on principal component analysis of plane triangles |
| CN105371870A (en) * | 2015-11-27 | 2016-03-02 | 北京空间飞行器总体设计部 | Star map data based method for measurement of in-orbit precision of star sensor |
| CN109238269A (en) * | 2018-09-05 | 2019-01-18 | 南京航空航天大学 | A kind of star pattern matching method across the fusion of discontinuous visual field |
| CN109359577A (en) * | 2018-10-08 | 2019-02-19 | 福州大学 | A kind of Complex Background number detection system based on machine learning |
| CN109813303A (en) * | 2019-03-08 | 2019-05-28 | 北京航空航天大学 | A kind of method for recognising star map for not depending on calibrating parameters based on the ballot of angle mould formula cluster |
Non-Patent Citations (1)
| Title |
|---|
| 姜笛: "基于最优路径的多视场全天自主星图识别", 《系统工程与电子技术》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114252088A (en) * | 2020-09-25 | 2022-03-29 | 南京理工大学 | Star map matching semi-physical verification method based on collimator |
| CN114252088B (en) * | 2020-09-25 | 2024-05-24 | 南京理工大学 | Star map matching semi-physical verification method based on collimator |
| CN112665579A (en) * | 2020-12-01 | 2021-04-16 | 中国人民解放军国防科技大学 | Star map identification method and device based on geometric verification |
| CN112665579B (en) * | 2020-12-01 | 2024-02-27 | 中国人民解放军国防科技大学 | Star map identification method and device based on geometric verification |
| CN112529809A (en) * | 2020-12-15 | 2021-03-19 | 航天信息股份有限公司 | Starry sky image color difference correction method and device, storage medium and aerospace observation equipment |
| CN112529809B (en) * | 2020-12-15 | 2024-01-02 | 航天信息股份有限公司 | Star image chromatic aberration correction method, device, storage medium and aerospace observation equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110793529B (en) | 2022-06-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111862126B (en) | Non-cooperative target relative pose estimation method combining deep learning and geometric algorithm | |
| CN110793529B (en) | Quick matching star map identification method | |
| Xiang et al. | Automatic registration of optical and SAR images via improved phase congruency model | |
| CN110245678B (en) | Image matching method based on heterogeneous twin region selection network | |
| CN112434745B (en) | Occlusion target detection and identification method based on multi-source cognitive fusion | |
| CN113706482A (en) | High-resolution remote sensing image change detection method | |
| CN110246181B (en) | Anchor point-based attitude estimation model training method, attitude estimation method and system | |
| CN109034184B (en) | Grading ring detection and identification method based on deep learning | |
| CN109785371A (en) | A kind of sun image method for registering based on normalized crosscorrelation and SIFT | |
| CN113963240B (en) | Comprehensive detection method for multi-source remote sensing image fusion target | |
| RU2476825C2 (en) | Method of controlling moving object and apparatus for realising said method | |
| CN112184577A (en) | Single image defogging method based on multi-scale self-attention generation countermeasure network | |
| CN113838064B (en) | Cloud removal method based on branch GAN using multi-temporal remote sensing data | |
| CN112700489B (en) | Ship-based video image sea ice thickness measurement method and system based on deep learning | |
| CN111414954A (en) | Rock image retrieval method and system | |
| CN115100545A (en) | Target detection method for small parts of failed satellite under low illumination | |
| CN113538347A (en) | Image detection method and system based on efficient bidirectional path aggregation attention network | |
| CN111860651A (en) | Monocular vision-based semi-dense map construction method for mobile robot | |
| CN117036300A (en) | Road surface crack identification method based on point cloud-RGB heterogeneous image multistage registration mapping | |
| CN112634130A (en) | Unmanned aerial vehicle aerial image splicing method under Quick-SIFT operator | |
| CN111626241A (en) | Face detection method and device | |
| CN114066795A (en) | DF-SAS high-low frequency sonar image fine registration fusion method | |
| CN113239828A (en) | Face recognition method and device based on TOF camera module | |
| CN117151983A (en) | Image full-color sharpening method based on wavelet heuristics and high-frequency enhancement | |
| CN112669452A (en) | Object positioning method based on convolutional neural network multi-branch structure |
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 |