CN108507564A - A kind of star sensor method for positioning mass center based on point spread function fitting - Google Patents

A kind of star sensor method for positioning mass center based on point spread function fitting Download PDF

Info

Publication number
CN108507564A
CN108507564A CN201810698182.2A CN201810698182A CN108507564A CN 108507564 A CN108507564 A CN 108507564A CN 201810698182 A CN201810698182 A CN 201810698182A CN 108507564 A CN108507564 A CN 108507564A
Authority
CN
China
Prior art keywords
asterism
picture
asterism picture
target
spread function
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
Application number
CN201810698182.2A
Other languages
Chinese (zh)
Other versions
CN108507564B (en
Inventor
曹阳
李保权
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.)
National Space Science Center of CAS
Original Assignee
National Space Science Center of CAS
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 National Space Science Center of CAS filed Critical National Space Science Center of CAS
Priority to CN201810698182.2A priority Critical patent/CN108507564B/en
Publication of CN108507564A publication Critical patent/CN108507564A/en
Application granted granted Critical
Publication of CN108507564B publication Critical patent/CN108507564B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/02Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by astronomical means
    • G01C21/025Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by astronomical means with the use of startrackers

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Astronomy & Astrophysics (AREA)
  • Automation & Control Theory (AREA)
  • General Physics & Mathematics (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses it is a kind of based on point spread function fitting star sensor method for positioning mass center, the method includes:Step 1) obtains a width and refers to asterism picture, calculates the barycenter with reference to asterism picture;Step 2) obtains effective point spread function with reference to asterism picture using sinc function pairs into row interpolation, to obtain any position asterism picture resampling;After step 3) is for a width target asterism picture, a square zoning is selected centered on asterism pixel as where peak value, then by target asterism as in zoning part and step 2) in obtained resampling asterism picture carry out least square fitting, target asterism picture is estimated relative to the centroid motion with reference to asterism picture;Step 4) sums the target asterism picture that the barycenter for the reference asterism picture that step 1) obtains and step 3) obtain relative to the estimation of centroid motion and the zoning top left co-ordinate of target asterism picture for referring to asterism picture, so that it may to obtain the barycenter of target asterism picture.

Description

A kind of star sensor method for positioning mass center based on point spread function fitting
Technical field
The present invention relates to spacecraft attitude fields of measurement, more particularly to carry out the star sensor skill of attitude measurement based on star chart A kind of art, and in particular to star sensor method for positioning mass center based on point spread function fitting.
Background technology
Star sensor is a kind of equipment carrying out attitude measurement as object of reference using fixed star, has precision height, without cumulative errors Advantage, be widely used in aerospace field.In recent years, with the fast development in the fields such as earth observation, space science, It is higher and higher to space attitude control and the required precision of attitude measurement.Therefore the measurement accuracy for improving star sensor is always sky Between technical field lasting pursuit.
The center coordination precision for improving asterism picture is to determine the key factor of attitude measurement accuracy.Traditional centroid method has calculation The advantages of method is simple, calculating speed is fast, strong robustness is that the asterism image quality heart most commonly seen in star sensor field at present is fixed Position method.One defect of traditional centroid method is that its systematic error is larger, although dimensional Gaussian point spread function pair can be used Systematic error is analyzed and compensated can reduce systematic error to a certain extent, but the defocus that general star sensor uses The point spread function of optical system and the shape difference of dimensional Gaussian point spread function are larger, especially optical aberration compared with Compensation effect unobvious when big.In addition, the photon numbers that the pixel of barycenter window edge detects are considerably less, traditional centroid method is but given Their higher weights are given, this causes the noise resisting ability of traditional centroid method poor.It is not deposited then based on point spread function fitting In above-mentioned two problems, ultimate challenge that such methods face is that the point spread function of optical system is difficult accurately to learn, at present In star sensor field, there are no a kind of method for positioning mass center based on point spread function fitting.
Invention content
It is an object of the invention to overcome conventional star sensor method for positioning mass center measurement accuracy it is low, by optical aberration shadow Ring big, noise resisting ability difference defect.
To achieve the goals above, the present invention proposes a kind of star sensor center coordination being fitted based on point spread function Method, the method includes:
Step 1) obtains a width and refers to asterism picture, calculates the barycenter with reference to asterism picture;
Step 2) obtains effective point spread function with reference to asterism picture using sinc function pairs into row interpolation, arbitrary to obtain The resampling of the asterism picture of position;
After step 3) is for a width target asterism picture, selection one is square centered on asterism pixel as where peak value Zoning, then by target asterism as in zoning part and step 2) in obtained resampling asterism picture carry out it is minimum Two multiply fitting, are estimated relative to the centroid motion with reference to asterism picture target asterism picture;
The target asterism picture that step 4) obtains the barycenter for the reference asterism picture that step 1) obtains and step 3) is relative to ginseng The zoning top left co-ordinate of the estimation and target asterism picture of examining the centroid motion of asterism picture is summed, so that it may to obtain The barycenter of target asterism picture.
As a kind of improvement of the above method, the step 1) specifically includes:
Step 1-1) allow star sensor point spread function on detector array surface along x-axis and y axis using high precision turntable It is moved with moving step length a/M, multiframe asterism picture is acquired on the position after each movement and is averagely obtained each position Asterism data;A is Pixel Dimensions, and M is to move step number along x-axis and y-axis both direction, amounts to mobile M × M position;
Step 1-2) to step 1-1) in the asterism data of each position that obtain pre-process, the pretreatment includes Details in a play not acted out on stage, but told through dialogues calibration and flat field calibration, the asterism picture after being pre-processed;
Step 1-3) in the asterism picture after each pretreatment, pixel of the asterism as where peak value is found, and with asterism A square zoning is selected centered on the pixel as where peak value;
Step 1-4) to the asterism picture of each position in step 1-3) part in the zoning of selection carries out group again It closes, obtains a width and refer to asterism picture;
Wherein,It is asterism picture when point spread function moves along the x-axis i steps, moves along y-axis j steps in zoning M rows the n-th row pixel value, 1≤m≤n_roi, 1≤n≤n_roi, n_roi be square zoning the length of side, m It is positive integer with n;
Step 1-5) calculate the barycenter for referring to asterism picture:
Wherein,To refer to the barycenter of asterism picture,It is the pth of the reference asterism picture reconfigured The value of row q row pixels, the pixel sampling interval with reference to asterism picture are a/M, 1≤p≤M × nroi, 1≤q≤M × nroi, p, q For positive integer.
As a kind of improvement of the above method, the value range of the M is:4~10.
As a kind of improvement of the above method, the length of side n_roi of the zoning of the square is 5 or 7.
As a kind of improvement of the above method, the resampling of the step 2) is:
Wherein,It is that barycenter offsets by (Δ x relative to reference to asterism picturec,Δyc) resampling Asterism picture.
As a kind of improvement of the above method, the centroid motion in the step 3) is estimated as:
Wherein,It is in the zoning of square of the target asterism picture centered on asterism pixel as where peak value M rows the n-th row pixel value,It is target asterism picture and the estimated value with reference to centroid motion between asterism picture.
As a kind of improvement of the above method, the centroid estimation of target asterism picture is in the step 4):
Wherein,It is counting for square of the target asterism picture centered on asterism pixel as where peak value The coordinate value of the top left corner pixel in region is calculated,It is the centroid estimation of target asterism picture.
The advantage of the invention is that:
In conjunction with the modulation transfer function and image sensor pixel frequency response characteristic of optical system of star sensor, the present invention Propose a kind of star sensor method for positioning mass center based on asterism as resampling;Not only center coordination precision is high for this method, and And influenced by optical aberration small, it can be effectively improved in the case where not increasing existing star sensor resource requirement The measurement accuracy of star sensor.
Description of the drawings
Fig. 1 is the flow chart for the star sensor method for positioning mass center of the present invention being fitted based on point spread function.
Specific implementation mode
In conjunction with attached drawing, the invention will be further described.
With reference to figure 1, the star sensor method for positioning mass center of the invention based on point spread function fitting includes the following steps:
Step 1) using high precision turntable allows star sensor point spread function in detector array (CCD in the lab Or APS) surface moved along x-axis and y-axis with a fixed step size, so that moving range is covered 1 pixel, in each station acquisition Multiframe asterism picture is averaged.
Step 2) pre-processes the asterism data of each position of acquisition in step 1), and the pretreatment includes dark Field calibration and flat field calibration, the asterism picture after being pre-processed.
Step 3), step 2) obtain initial position pretreatment after asterism picture in, find asterism as peak value institute Pixel, and by asterism as selecting a square zoning centered on pixel where peak value.
Step 4), the part to the asterism picture of each position obtained in step 2) in the zoning that step 3) obtains It is reconfigured, obtains the very low reference asterism picture in a width pixel sampling interval.
Step 5), the barycenter that the reference asterism picture obtained in step 4) is calculated using traditional centroid method.
Step 6), using the reference asterism picture obtained in step 4), accurately had into row interpolation using sinc functions Imitate point spread function, so as to obtain any position asterism picture resampling.
Step 7), after obtaining a width target asterism picture, equally by asterism as selection one is being just centered on pixel where peak value Rectangular zoning, then by target asterism as in zoning part and step 6) in obtained resampling asterism picture into Row least square fitting estimates target asterism picture relative to the centroid motion with reference to asterism picture.
It is step 8), the target asterism picture obtained in the barycenter of the reference asterism picture obtained in step 5) and step 7) is opposite It sums in the estimation of centroid motion of reference asterism picture and the zoning top left co-ordinate of target asterism picture, so that it may with Obtain the barycenter of target asterism picture.
Each step in the method for the present invention is described further below.
In step 1), the step number that moves along x-axis and y-axis both direction is M, and moving step length a/M, wherein a are picture Plain size amounts to mobile M × M position.The selection of M is related with camera lens F numbers, the size of defocusing amount, in order to ensure point spread function Number reconstruction accuracy, reduces operand, generally takes 4~10.
It is related with noise of detector, background noise level etc. in the frame number of each station acquisition asterism picture in step 1), In order to reduce the influence of noise, general frame number should be greater than 10.
In step 3), the size of square zoning, as size determination, can be 5 × 5,7 × 7 etc. according to asterism. The selection of window size, if crossing conference influences calculating speed;And if too small, can error be generated due to signal cutout.
In step 4), mode such as following formula (1) that specific asterism picture reconfigures:
WhereinIt is asterism when point spread function moves along the x-axis i steps, moves along y-axis j steps as in zoning The value range of the value of m rows the n-th row pixel, m, n is from 1 to n_roi, and n_roi is the size of zoning;
In step 5), it is with reference to the specific method of the asterism image quality heart using the calculating of traditional centroid method:
WhereinIt is the barycenter for the reference asterism picture being calculated,It is the reference star reconfigured The value of the pth row q row pixels of point picture, the pixel sampling interval with reference to asterism picture is a/M, and the value range of p, q are from 1 to M* n_roi。
Due to average by the multiframe asterism picture in step 1), the noise with reference to asterism picture is very low, so with reference to asterism picture The random error very little of centroid calculation result.Very low additionally, due to the pixel sampling interval with reference to asterism picture, result of calculation is Error of uniting also can be very low.
In step 6), effective point spread function is the convolution of optical system point spread function and pixel response function, star Point seems the sampling of effective point spread function, and the sampling interval is Pixel Dimensions a.
In step 6), emulate and the experimental results showed that the high fdrequency component of effective point spread function is very low.Therefore, although With reference to asterism as sampling interval a/M cannot meet the requirement of sampling thheorem, expanded with sinc function interpolations reconstruct available point Dissipating the precision of function can meet the requirements.
In step 6), the method such as following formula (3) of the resampling of any position asterism picture is specifically obtained by interpolation:
In formulaIt is that barycenter offsets by (Δ x relative to reference to asterism picturec,Δyc) resampling Asterism picture.
In step 7), the size shape of zoning and being consistent in step 3).
The specific centroid motion method of estimation such as following formula (4) in step 7):
In formulaIt is the value of m row n-th row pixel of the target asterism picture in zoning,It is target The estimated value of centroid motion between asterism picture and reference asterism picture.
In step 8), the method for the specific barycenter for calculating target asterism picture is as follows:
In formulaIt is the coordinate value of the zoning top left corner pixel selected in step 7),It is the centroid estimation of target asterism picture.
It should be noted last that the above examples are only used to illustrate the technical scheme of the present invention and are not limiting.Although ginseng It is described the invention in detail according to embodiment, it will be understood by those of ordinary skill in the art that, to the technical side of the present invention Case is modified or replaced equivalently, and without departure from the spirit and scope of technical solution of the present invention, should all be covered in the present invention Right in.

Claims (7)

1. a kind of star sensor method for positioning mass center based on point spread function fitting, the method includes:
Step 1) obtains a width and refers to asterism picture, calculates the barycenter with reference to asterism picture;
Step 2) obtains effective point spread function with reference to asterism picture using sinc function pairs into row interpolation, to obtain any position Asterism picture resampling;
After step 3) is for a width target asterism picture, a square calculating is selected centered on asterism pixel as where peak value Region, then by target asterism as in zoning part and step 2) in obtained resampling asterism picture carry out least square Target asterism picture is estimated in fitting relative to the centroid motion with reference to asterism picture;
The target asterism picture that step 4) obtains the barycenter for the reference asterism picture that step 1) obtains and step 3) is relative to reference star The estimation of centroid motion of point picture and the zoning top left co-ordinate of target asterism picture are summed, so that it may to obtain target The barycenter of asterism picture.
2. the star sensor method for positioning mass center according to claim 1 based on point spread function fitting, which is characterized in that The step 1) specifically includes:
Step 1-1) using high precision turntable allow star sensor point spread function on detector array surface along x-axis and y-axis to move Dynamic step-length a/M is moved, and the star that multiframe asterism picture is averagely obtained each position is acquired on the position after each movement Point data;A is Pixel Dimensions, and M is to move step number along x-axis and y-axis both direction, amounts to mobile M × M position;
Step 1-2) to step 1-1) in the asterism data of each position that obtain pre-process, the pretreatment includes details in a play not acted out on stage, but told through dialogues Calibration and flat field calibration, the asterism picture after being pre-processed;
Step 1-3) in the asterism picture after each pretreatment, pixel of the asterism as where peak value is found, and with asterism as peak A square zoning is selected centered on pixel where value;
Step 1-4) to the asterism picture of each position in step 1-3) part in the zoning of selection reconfigures, obtains Asterism picture is referred to a width;
Wherein,It is that point spread function moves along the x-axis i steps, moves m of the asterism picture in zoning when j is walked along y-axis The value of row the n-th row pixel, 1≤m≤n_roi, 1≤n≤n_roi, n_roi are the length of sides of the zoning of square, and m and n are Positive integer;
Step 1-5) calculate the barycenter for referring to asterism picture:
Wherein,To refer to the barycenter of asterism picture,It is the pth row q of the reference asterism picture reconfigured The value of row pixel, the pixel sampling interval with reference to asterism picture are a/M, 1≤p≤M × nroi, 1≤q≤M × nroi, p, q are just whole Number.
3. the star sensor method for positioning mass center according to claim 2 based on point spread function fitting, which is characterized in that The value range of the M is:4~10.
4. the star sensor method for positioning mass center according to claim 2 or 3 based on point spread function fitting, feature exist In the length of side n_roi of the zoning of the square is 5 or 7.
5. the star sensor method for positioning mass center according to claim 4 based on point spread function fitting, which is characterized in that The resampling of the step 2) is:
Wherein,It is that barycenter offsets by (Δ x relative to reference to asterism picturec, Δ yc) resampling asterism Picture.
6. the star sensor method for positioning mass center according to claim 5 based on point spread function fitting, which is characterized in that Centroid motion in the step 3) is estimated as:
Wherein,It is the m in the zoning of square of the target asterism picture centered on asterism pixel as where peak value The value of row the n-th row pixel,It is target asterism picture and the estimated value with reference to centroid motion between asterism picture.
7. the star sensor method for positioning mass center according to claim 6 based on point spread function fitting, which is characterized in that The centroid estimation of target asterism picture is in the step 4):
Wherein,It is the calculating area of square of the target asterism picture centered on asterism pixel as where peak value The coordinate value of the top left corner pixel in domain,It is the centroid estimation of target asterism picture.
CN201810698182.2A 2018-06-29 2018-06-29 Star sensor centroid positioning method based on point spread function fitting Active CN108507564B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810698182.2A CN108507564B (en) 2018-06-29 2018-06-29 Star sensor centroid positioning method based on point spread function fitting

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810698182.2A CN108507564B (en) 2018-06-29 2018-06-29 Star sensor centroid positioning method based on point spread function fitting

Publications (2)

Publication Number Publication Date
CN108507564A true CN108507564A (en) 2018-09-07
CN108507564B CN108507564B (en) 2020-06-16

Family

ID=63404034

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810698182.2A Active CN108507564B (en) 2018-06-29 2018-06-29 Star sensor centroid positioning method based on point spread function fitting

Country Status (1)

Country Link
CN (1) CN108507564B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109443381A (en) * 2018-10-17 2019-03-08 北京遥感设备研究所 A kind of star sensor Centroid accuracy self-adapting compensation method
CN111854736A (en) * 2020-06-11 2020-10-30 中国科学院西安光学精密机械研究所 Error-suppression star point centroid positioning method
CN113487740A (en) * 2021-05-21 2021-10-08 北京控制工程研究所 Space target nanometer precision imaging positioning method

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6373041B1 (en) * 2000-02-18 2002-04-16 Godrich Corporation Image tracker having means for eliminating systematic error in centroid determination of reconstructed images
CN101013065A (en) * 2006-03-21 2007-08-08 北京航空航天大学 Pixel frequency based star sensor high accuracy calibration method
CN103162711A (en) * 2013-02-28 2013-06-19 北京航空航天大学 Method and device for high-dynamic star sensor image intensifier error compensation
CN103323027A (en) * 2013-05-30 2013-09-25 北京控制工程研究所 Star point reconstruction-based star sensor dynamic-compensation method
CN105023281A (en) * 2015-05-25 2015-11-04 中国科学院空间科学与应用研究中心 Method for computing center of mass of star map based on point spread function wave front correction
CN106570902A (en) * 2016-11-04 2017-04-19 中国科学院国家空间科学中心 PSF relative centroid calculation method acquired based on detector pixel response spectrum

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6373041B1 (en) * 2000-02-18 2002-04-16 Godrich Corporation Image tracker having means for eliminating systematic error in centroid determination of reconstructed images
CN101013065A (en) * 2006-03-21 2007-08-08 北京航空航天大学 Pixel frequency based star sensor high accuracy calibration method
CN100376883C (en) * 2006-03-21 2008-03-26 北京航空航天大学 Pixel frequency based star sensor high accuracy calibration method
CN103162711A (en) * 2013-02-28 2013-06-19 北京航空航天大学 Method and device for high-dynamic star sensor image intensifier error compensation
CN103323027A (en) * 2013-05-30 2013-09-25 北京控制工程研究所 Star point reconstruction-based star sensor dynamic-compensation method
CN105023281A (en) * 2015-05-25 2015-11-04 中国科学院空间科学与应用研究中心 Method for computing center of mass of star map based on point spread function wave front correction
CN106570902A (en) * 2016-11-04 2017-04-19 中国科学院国家空间科学中心 PSF relative centroid calculation method acquired based on detector pixel response spectrum

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
MIGHELL K J ET.AL: "Stellar photometry and astrometry with discrete point spread function", 《MONTHLY NOTICES OF THE ROYAL ASTRONOMICAL SOCIETY》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109443381A (en) * 2018-10-17 2019-03-08 北京遥感设备研究所 A kind of star sensor Centroid accuracy self-adapting compensation method
CN109443381B (en) * 2018-10-17 2022-05-20 北京遥感设备研究所 Star sensor centroid precision self-adaptive compensation method
CN111854736A (en) * 2020-06-11 2020-10-30 中国科学院西安光学精密机械研究所 Error-suppression star point centroid positioning method
CN113487740A (en) * 2021-05-21 2021-10-08 北京控制工程研究所 Space target nanometer precision imaging positioning method

Also Published As

Publication number Publication date
CN108507564B (en) 2020-06-16

Similar Documents

Publication Publication Date Title
CN108507564A (en) A kind of star sensor method for positioning mass center based on point spread function fitting
CN109146794A (en) A kind of light field image rotation error bearing calibration
CN106971408B (en) A kind of camera marking method based on space-time conversion thought
Aykin et al. On feature extraction and region matching for forward scan sonar imaging
CN108615244A (en) A kind of image depth estimation method and system based on CNN and depth filter
CN105182678B (en) A kind of system and method based on multichannel camera observation space target
CN113012234B (en) High-precision camera calibration method based on plane transformation
EP2901236A1 (en) Video-assisted target location
CN109724586A (en) A kind of spacecraft relative pose measurement method of fusion depth map and point cloud
CN109084959B (en) Optical axis parallelism correction method based on binocular distance measurement algorithm
CN108562900B (en) SAR image geometric registration method based on elevation correction
CN107360377B (en) Vehicle-mounted video image stabilization method
CN109270527A (en) Circular track SAR sub-aperture path image sequence combines correlation DEM extracting method
CN110595374A (en) Large structural part real-time deformation monitoring method based on image transmission machine
CN110849354A (en) Star point extraction and compensation method under condition of last life stage of star sensor
CN107301628B (en) It is trembled image deblurring method based on trembling as moving the satellite platform of track
CN104156933A (en) Image registering method based on optical flow field
JP5267100B2 (en) Motion estimation apparatus and program
CN111127334B (en) SAR image real-time geometric correction method and system based on RD plane pixel mapping
CN112419427A (en) Method for improving time-of-flight camera accuracy
CN113689474B (en) Tracking method based on gaze tracking one-dimensional high-speed moving target system
CN108399629A (en) A kind of pattern matrix light stream method of estimation of artificial compound eye camera
CN113916136A (en) High-rise structure dynamic displacement measurement method based on unmanned aerial vehicle aerial photography
CN113552648A (en) Astronomical positioning and automatic tracking method and system for space target and electronic equipment
CN109493382B (en) Fixed star high-precision position extraction method based on intra-pixel response

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