WO1998036373A1 - Procede pour reconnaitre dans un catalogue preetabli des etoiles detectees par un senseur stellaire - Google Patents
Procede pour reconnaitre dans un catalogue preetabli des etoiles detectees par un senseur stellaire Download PDFInfo
- Publication number
- WO1998036373A1 WO1998036373A1 PCT/FR1998/000277 FR9800277W WO9836373A1 WO 1998036373 A1 WO1998036373 A1 WO 1998036373A1 FR 9800277 W FR9800277 W FR 9800277W WO 9836373 A1 WO9836373 A1 WO 9836373A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- stars
- class
- doublets
- classes
- star
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S3/00—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received
- G01S3/78—Direction-finders for determining the direction from which infrasonic, sonic, ultrasonic, or electromagnetic waves, or particle emission, not having a directional significance, are being received using electromagnetic waves other than radio waves
- G01S3/782—Systems for determining direction or deviation from predetermined direction
- G01S3/785—Systems for determining direction or deviation from predetermined direction using adjustment of orientation of directivity characteristics of a detector or detector system to give a desired condition of signal derived from that detector or detector system
- G01S3/786—Systems for determining direction or deviation from predetermined direction using adjustment of orientation of directivity characteristics of a detector or detector system to give a desired condition of signal derived from that detector or detector system the desired condition being maintained automatically
- G01S3/7867—Star trackers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/757—Matching configurations of points or features
Definitions
- the invention relates to a method for recognizing all or part of the stars detected by a stellar sensor in a pre-established catalog of stars of the stellar vault, for a user processing system of stellar recognition.
- a stellar sensor in a pre-established catalog of stars of the stellar vault, for a user processing system of stellar recognition.
- it is for example possible to estimate the attitude of the reference of the stellar sensor or to refine an estimate also using other means, which then makes it possible to go back to the attitude of the vehicle, for example a aircraft or submarine, on which the stellar sensor is installed.
- the invention aims to improve the recognition rate of known methods and to minimize the calculation time.
- This is achieved according to the present invention by means of a method as defined above and characterized in that: stars of the catalog are grouped according to classes, considering that two stars belong to the same class if and only if they are confused or if there is a path which connects them, a path being a sequence of possible doublets such that any pair of consecutive doublets of this sequence has at least one star in common, - lists are created class by class by matching all or part of the stars of a class with stars detected to constitute a list and
- the invention thus makes it possible to create a set of sub-catalogs such that the stars actually present in the field of view have a tiny chance of being dispersed in several of these sub-catalogs whose stars have a particularly great chance of having were present in the field of view, those which are too little linked having been evacuated by purification.
- the method of the invention leads to the fact that the pairings are carried out on the basis of purified sub-catalogs of very small size, which most of the time largely compensates for the increase in computation time induced by the multiplicity of sub-catalogs. .
- the subject of the invention is also a pairing method by which, with equal performance of the star sensor, the recognition rate is significantly improved compared to known methods.
- Modes of implementation of the method of the invention will be described below, and the description and the figures will show other advantageous characteristics of these modes of implementation. In the figures:
- FIG. 1 is a diagram showing detected and possible doublets in a characterization space
- - Figure 2 is a diagram showing possible doublets in a representation of the sky
- - Figure 3 is a diagram of a list of possible doublets formed from a list of detected doublets
- FIG. 4 is a diagram of the celestial vault in which the field of view of the sensor has been represented and, for different classes, the paths which connect the stars of each class;
- FIG. 5 is a diagram representing a method of classifying;
- FIG. 6 is a diagram representing a pairing process, and
- FIG. 7 is a diagram of implementation of a method for eliminating pairs of paired stars which are outliers.
- a preliminary step to the method of the invention consists in reconstructing, from the raw measurement of the sensor, a set of stars supposed to have been present at the time of the measurement.
- the output from this step is a list of "stars captured". These stars are often described by their magnitude and their position in a reference frame (for example that of the sensor).
- the next step or first step of the method of the invention consists in keeping only part of the "captured stars" for the rest of the treatments. For example, we suggest that the K stars of lowest magnitude (i.e. the brightest), K being an integer chosen in advance.
- the stars from this selection are called “detected stars”.
- the second step of the process consists in coupling the detected stars two by two into “detected doublets”.
- the third step of the method possibly carried out as all or part of the set of detected doublets is searched, aims to list all or part of the "possible doublets", that is to say doublets of the catalog of stars that are likely to have been present in the field of view during the measurement.
- the interest of working on the difference of magnitudes is as follows: errors on the magnitude in the form of bias are automatically eliminated.
- the neighborhood is for example defined by an ellipsoid or a parallelepiped, and it can be centered on the detected doublet.
- this step you can only collect the doublets that are in certain areas of the sky known as those that the sensor is likely to target. This can be achieved by dynamically creating, before this step, a sub-catalog of stars and / or doublets replacing the complete catalogs. This can also be achieved by a test when passing over a doublet, during the course of all the doublets corresponding to the stars in the catalog of stars or doublets.
- Figure 2 is a diagram of the sky where an example of possible doublets from the catalog has been shown.
- the set of possible doublets is called "D".
- the set of corresponding possible stars is implicitly associated with it.
- Another step consists according to the invention in grouping the stars of D into classes.
- a path h as a sequence ⁇ 0 , ..., ⁇ N _, of elements of D, such that any pair of doublets consecutive in this suite has at least one common star.
- This equivalence relation defines in the star catalog equivalence classes.
- a class is in the following the set of stars corresponding to an equivalence class and / or the set of corresponding possible doublets.
- Figure 4 which represents stars from the catalog, the classes are identified by a curved loop.
- FIG. 5 which presents a method of classifying, starting from a star belonging to one of the possible doublets and which constitutes the first star of a new class, we relate gradually to this classifies stars which are connected by at least one path to the first star.
- the recursive process must be stopped. To do this, we can “mark” the stars already processed so that they are no longer treat again. We can also mark the doublets so as not to treat them a second time, when we access them from the other end.
- a class as: list of objects pointing to the stars in the catalog or reproducing these stars. - list of objects pointing to the complete list of possible doublets or reproducing these doublets (a class then being represented by the list of doublets containing its stars),
- Virtual classes can also be carried out. Contrary to the previous “materialized classifications” (consisting for example of creating lists or setting up pointers in IT structures), virtual classifications consist to attribute to each star certain characteristics of the equivalence class to which it belongs on the basis of existing possible doublets. These characteristics can be the number of stars and / or the number of possible doublets of the group.
- the recursive method above makes it possible to carry out a virtual classification without creating a list provided that it is applied twice, the first time to evaluate the characteristics of the equivalence class and the second to assign these characteristics to each star. of the group.
- all or some of the possible stars are subjected to a purification. This consists in permanently deleting from the list or lists in which they appear, on the basis of a criterion, certain possible stars as well as all the possible doublets containing them.
- This criterion consists, for example, in rejecting a star that is too close to another star in the sense of a standard defined in a space for characterizing doublets, or in rejecting a star if the number of possible doublet connections between this star and the other stars is below a ' threshold.
- this threshold may depend on the characteristics of the last class to which the star considered belonged (an important characteristic being the number of stars in this class). Applying this criterion to all the stars of a group of stars to be treated - as the case may be and according to the times to all the possible stars or to a part of them corresponding for example to a class - constitutes a purge pass.
- Consecutive passes can be carried out on the basis of the same criterion, in predefined number or until no more stars are deleted. When a pass has led to no stars being deleted, there is no point in making the following passes for the same episode.
- the whole of the purification consists of several successions of episodes, respectively applied to various groups of stars made up of the complete set of possible stars, of a class, or of the residue of stars having previously formed a class.
- Each episode can be defined by a different purification criterion.
- the number of episodes can however be possibly reduced to zero, the purification is then nonexistent.
- Any purification pass which would be applied to a class is likely not only to reduce the size of this class by deleting one or more stars, but also to transform the structure of stars and possible doublets which constituted the class into a structure no longer having the properties of an equivalence class.
- the characteristics of the classes intervene in the criterion of a later pass we can use the characteristics of the last class to which the star considered belonged. If the number of stars of the last class intervenes in the criterion, one can however use in the criterion not the number of stars of the origin class, but this number reduced by one unit with each deletion of star.
- the classification operations and the production of episodes can be advantageously articulated, so as to refine the purification criterion thanks to the characteristics of the group of stars (an equivalence class which for the process is not yet a class) to which the star considered is linked by possible doublets.
- Classifications can be performed after certain episodes.
- the succession of episodes and classifications is constructed from one or more "episode tables" such as that presented in Table I.
- each episode is characterized by its number of passes and by its criterion (in this example the criterion is defined by the minimum number of possible doublets containing a star that we keep, depending on the number of stars in the group).
- the last episode of any episode table is necessarily followed by a classification of the stars produced by the last episode (except, possibly, if this group of stars has remained unchanged through the last episodes applied to the stars of the group since their last classification). If this last classification is initializing and if several classes are produced then we resume the purification at the first episode of a new table of episodes. If only one class is produced or if the classification is not initializing then the purification process stops for the corresponding stars and the classes produced (possibly reduced to only one) are final.
- these three processes require that over the purification of a group of stars, information is given of the attributes of this group describing its current purification point (reference from the episode table, reference from the episode current inside the table and reference of the current pass inside the current episode). The first two of these attributes may need to be kept when classifying so that you can determine which next episode will be applied to the resulting classes.
- the purification criterion involves the characteristics of the group of stars containing the treated star (for example the number of stars and / or the number of links between this star and the other stars of the class), there are two possibilities. Either a first course of the group of stars submitted to the first pass pass is made to mark the possible stars and the possible doublets to be eliminated. The criterion is then each time calculated on the basis of the initial characteristics of the group, eliminations being carried out in the different lists in a second step. Either eliminations are carried out progressively, with the disadvantage that the characteristics of the group can change as successive calculations of the purification criteria (following the deletion of elements). In all cases, classes with too few stars can be eliminated as all classifications are carried out.
- the different classes constitute as many sub-catalogs. These offer the following properties: on the one hand, there is little chance that the stars actually present in the field of view of the sensor during the measurement will be distributed between several classes. Indeed, a good part of the detected doublets connecting the detected stars have led to a correct selection of the possible doublets between the good possible stars (not yet known). The numerous connections between these .stars will have prevented them from being separated into several classes during the classification-purification operations. On the other hand, in the class effectively containing the stars present in the field of view during the measurement, the bad stars have a good chance of having been eliminated by the purification.
- the classes can therefore be treated independently, and each time an attempt is made to match the stars detected with the possible stars of the class considered. Any known process or combination of processes known can be used for this. For each class thus treated, a “list of paired star pairs” is obtained.
- the next step in the process is therefore the matching of the possible stars (of each class successively) with the detected stars resulting from the stellar measurement.
- this detected star has already been paired with a star of the current class, we can move on to the next star of the class, after having possibly marked this class as being “inconsistent”.
- the purification process which has just been presented is one of these processes which can be implemented before the classification-purification operations, without loss of performance in the case where the purification is non-existent or very mild .
- the pairs of paired stars created progressively can be added to the “list of pairs of paired stars” corresponding to the classes as they are created. Otherwise, the lists are created after the classification.
- pairing attempts can be stopped when it is estimated that there are sufficient paired pairs. It is thus advantageous to treat the possible stars in the decreasing order of the confidence which one can grant them. The decreasing order of the number of possible doublets in which the possible treated stars are present is advised in the case of the process which has just been presented.
- the confidence values can depend on the following terms: the number N, of stars of the class. This number can intervene by the intermediary of the hyperbolic tangent term of the number of stars multiplied by a constant, the whole raised to a certain power.
- the average connection rate T of the class, defined as its number of possible doublets of the class divided by R, (IS j -1) / 2 (the maximum number of doublets taking into account the number of stars). This number can intervene through one of its powers.
- the a posteriori confidence value since it is calculated after the pairing, can be based on other terms such as the rate of paired stars (which is the ratio of the number of paired stars of the class on the total number of stars in the class), or on a value characterizing the inconsistencies that occurred during the pairing of the stars of the class.
- the a priori and a posteriori confidence values may have a product of the terms presented above and relating to them respectively.
- Various treatments can also be applied to reject unreliable classes. For example, we can definitively reject classes whose a priori confidence value is less than a threshold fixed in advance. After the matching, we can reject the classes whose a posteriori confidence value is lower than a threshold fixed in advance, or those which have led to inconsistencies during the matching.
- the present process treats the classes in descending order of their a priori confidence index and stops the processing when one or more paired lists considered as sufficiently reliable have been obtained because their confidence index exceeds a threshold value.
- the possible gain in computation time is improved if the pairings are made for a class immediately after its final creation within the process of classifying-purifying. In this case, it is at each level of classification that the classes are processed in the order of their confidence index.
- a predetermined number of lists of paired star pairs can be selected, this number being however limited to the number of classes produced.
- each of the lists is advantageously accompanied by its a priori confidence criterion.
- N r be the number of these other pairs such that the difference is greater than a threshold value which may depend on the characteristics of the doublets, for example the average of their inter-star distances.
- the current torque is rejected if N r exceeds a limit possibly depending on the number of paired couples.
- the characterization space of the doublets can have as axes the inter-star distance, the sum of the magniments, the difference in magnitude (and no longer this time the absolute value of this difference).
- the space is such that the order in which we take the two stars is important, we start for the detected doublet and for the possible doublet by the stars of a pair paired. So that the criteria are applied on the basis of the initial characteristics of the list of couples, this process can advantageously be carried out in two passes: the first to mark the couples to be deleted and the second to effectively delete them from the lists.
- the method of deleting outliers can be applied before or after deleting lists on the basis
- the pairs of paired stars make it possible to restore the attitude or to improve such an estimate.
- the position of the stars detected on the sensor matrix makes it possible to go back to the direction of these stars in the reference frame linked to the sensor. Furthermore, the direction of the corresponding possible stars is known in the reference linked to the catalog. If we have at least two paired pairs, we can then, for example using a least squares algorithm, estimate the orientation of the sensor coordinate system in the catalog coordinate system. Since we know the orientation of the sensor reference in the satellite reference, we can then, by changing the reference, go back to the attitude of the satellite in the reference in the catalog and then in a reference, for example linked to the Earth.
- lists of objects stars, doublets, etc.
- This is the convention chosen to describe the process.
- the notion of "list” translates the existence of means allowing to materialize the links between a series of objects, allowing to find these objects to make them undergo treatments or to arrange them between them. This sequence is sometimes ordered and sometimes not ordered.
- these lists can take very different forms. In particular, they can be virtual, that is to say only materialized by markers (markers or pointers) in larger sets (for example the catalogs of stars and doublets).
- a characterization space of the three-dimensional doublets is defined, the axes of which are respectively relative to the angular distance, the difference and the sum of the magnitudes of the two stars of a doublet.
- a doublet is therefore represented in this space by a point, whether it is a detected doublet (corresponding to a pair of detected stars) or a possible doublet (ie a catalog doublet) .
- a classification and purification operation is then carried out based on Table I presented above.
- the purification criterion is therefore expressed as follows: a possible star is deleted if the number of possible doublets in which it is present is strictly less than a fixed value, depending from the second episode on the number of stars in the class.
- a possible star is deleted if the number of possible doublets in which it is present is strictly less than a fixed value, depending from the second episode on the number of stars in the class.
- the pairing operations are stopped when the number of paired stars is greater than 5.
- the posteriori confidence rate defined for class number j is calculated as follows:
- N j is the number of stars of the class
- T 3 the connection rate of the class
- ⁇ the rate of paired stars in the class (number of paired stars divided by the total number).
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Electromagnetism (AREA)
- Multimedia (AREA)
- Navigation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
Description
Claims
Priority Applications (6)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP53542398A JP2001519899A (ja) | 1997-02-13 | 1998-02-13 | 星センサーで検出された星を予め確立した星表中で同定する方法 |
EP98908183A EP0960382B9 (fr) | 1997-02-13 | 1998-02-13 | Procede pour reconnaitre dans un catalogue preetabli des etoiles detectees par un senseur stellaire |
RU99119312/09A RU99119312A (ru) | 1997-02-13 | 1998-02-13 | Способ распознавания звезд в предварительно составленном каталоге, обнаруженных астродатчиком |
DE69840021T DE69840021D1 (de) | 1997-02-13 | 1998-02-13 | Verfahren zur erkennung von durch einen sternsensor aufgefundenen sternen in einem vorerzeugten verzeichnis |
US09/367,428 US6324475B1 (en) | 1997-02-13 | 1998-02-13 | Method for identifying in a pre-established catalogue stars detected by a star sensor |
CA002280978A CA2280978A1 (fr) | 1997-02-13 | 1998-02-13 | Procede pour reconnaitre dans un catalogue preetabli des etoiles detectees par un senseur stellaire |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
FR97/01668 | 1997-02-13 | ||
FR9701668A FR2759474B1 (fr) | 1997-02-13 | 1997-02-13 | Procede pour reconnaitre dans un catalogue preetabli des etoiles detectees par un senseur stellaire |
Publications (1)
Publication Number | Publication Date |
---|---|
WO1998036373A1 true WO1998036373A1 (fr) | 1998-08-20 |
Family
ID=9503658
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/FR1998/000277 WO1998036373A1 (fr) | 1997-02-13 | 1998-02-13 | Procede pour reconnaitre dans un catalogue preetabli des etoiles detectees par un senseur stellaire |
Country Status (8)
Country | Link |
---|---|
US (1) | US6324475B1 (fr) |
EP (1) | EP0960382B9 (fr) |
JP (1) | JP2001519899A (fr) |
CA (1) | CA2280978A1 (fr) |
DE (1) | DE69840021D1 (fr) |
FR (1) | FR2759474B1 (fr) |
RU (1) | RU99119312A (fr) |
WO (1) | WO1998036373A1 (fr) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2831273A1 (fr) * | 2001-10-23 | 2003-04-25 | Thales Sa | Dispositif d'acquisition de cible, aeronef, systeme d'estimation de trajectoire et systeme de defense associes |
CN104819715A (zh) * | 2015-04-29 | 2015-08-05 | 西安电子科技大学 | 一种基于一维矢量模式的自主星识别方法 |
Families Citing this family (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3871891B2 (ja) * | 2001-02-27 | 2007-01-24 | 三菱電機株式会社 | 人工衛星の姿勢決定装置 |
US7007888B2 (en) * | 2003-11-25 | 2006-03-07 | The Boeing Company | Inertial position target measuring systems and methods |
US7310578B2 (en) * | 2004-01-09 | 2007-12-18 | The Boeing Company | Fast access, low memory, pair catalog |
CN100342213C (zh) * | 2004-12-28 | 2007-10-10 | 北京航空航天大学 | 一种基于三角形特征的星图识别方法 |
CN100357705C (zh) * | 2004-12-28 | 2007-12-26 | 北京航空航天大学 | 一种基于lpt变换的星图识别方法 |
US7382448B1 (en) | 2005-03-16 | 2008-06-03 | Celestron Acquisition, Llc | Alignment system for observation instruments |
US8222582B1 (en) * | 2008-09-30 | 2012-07-17 | Anderson Mark J | Celestial navigation using stellar narrow-band emission |
CN102072730B (zh) * | 2010-11-04 | 2012-11-21 | 北京航空航天大学 | 一种三角形星图识别方法 |
US20130006449A1 (en) * | 2011-06-30 | 2013-01-03 | George William Hindman | Apparatus, system and method for spacecraft navigation using extrasolar planetary systems |
CN102313549B (zh) * | 2011-07-28 | 2013-07-24 | 北京航空航天大学 | 一种基于惯性比特征的三角形星图识别方法 |
CN102410844B (zh) * | 2011-08-12 | 2013-12-11 | 北京航空航天大学 | 一种高动态星敏感器图像非均匀校正方法及装置 |
CN103335648B (zh) * | 2013-06-27 | 2015-11-11 | 北京航天自动控制研究所 | 一种自主星图识别方法 |
CN103954280B (zh) * | 2014-04-08 | 2016-09-21 | 北京控制工程研究所 | 一种快速和高鲁棒性自主恒星识别方法 |
CN103968833B (zh) * | 2014-04-30 | 2017-01-04 | 中国科学院长春光学精密机械与物理研究所 | 一种星图匹配前选取观测三角形的方法 |
KR101767197B1 (ko) * | 2016-05-09 | 2017-08-11 | 인하대학교 산학협력단 | 위성의 천측항법 알고리듬 검증을 위한 시험장치 및 이를 이용한 천측항법 알고리듬 검증방법 |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2691426B1 (fr) * | 1992-05-19 | 1997-09-19 | Aerospatiale | Procede et dispositif d'acquisition de la terre via la polaire pour satellite stabilise 3-axes en orbite de faible inclinaison. |
-
1997
- 1997-02-13 FR FR9701668A patent/FR2759474B1/fr not_active Expired - Fee Related
-
1998
- 1998-02-13 JP JP53542398A patent/JP2001519899A/ja active Pending
- 1998-02-13 RU RU99119312/09A patent/RU99119312A/ru not_active Application Discontinuation
- 1998-02-13 WO PCT/FR1998/000277 patent/WO1998036373A1/fr active IP Right Grant
- 1998-02-13 US US09/367,428 patent/US6324475B1/en not_active Expired - Lifetime
- 1998-02-13 CA CA002280978A patent/CA2280978A1/fr not_active Abandoned
- 1998-02-13 DE DE69840021T patent/DE69840021D1/de not_active Expired - Lifetime
- 1998-02-13 EP EP98908183A patent/EP0960382B9/fr not_active Expired - Lifetime
Non-Patent Citations (3)
Title |
---|
BEZOOIJEN VAN R W R: "AUTONOMOUS STAR TRACKER DEVELOPMENT", AUTOMATIC CONTROL IN AEROSPACE IFAC SYMPOSIUM, 17-21 JULY 1989, TSUKUBA, JP, 7 September 1992 (1992-09-07), pages 513 - 518, XP000645665 * |
GROTH E J: "A pattern-matching algorithm for two-dimensional coordinate lists", ASTRONOMICAL JOURNAL, MAY 1986, USA, vol. 91, no. 5, ISSN 0004-6256, pages 1244 - 1248, XP002044901 * |
SCHOLL M S: "SIX-FEATURE STAR-PATTERN IDENTIFICATION ALGORITHM", APPLIED OPTICS, vol. 33, no. 20, 10 July 1994 (1994-07-10), pages 4459 - 4464, XP000455186 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2831273A1 (fr) * | 2001-10-23 | 2003-04-25 | Thales Sa | Dispositif d'acquisition de cible, aeronef, systeme d'estimation de trajectoire et systeme de defense associes |
WO2003036323A1 (fr) * | 2001-10-23 | 2003-05-01 | Thales | Dispositif d'acquisition de cible, aeronef, systeme d'estimation de trajectoire et systeme de defense associes |
CN104819715A (zh) * | 2015-04-29 | 2015-08-05 | 西安电子科技大学 | 一种基于一维矢量模式的自主星识别方法 |
Also Published As
Publication number | Publication date |
---|---|
DE69840021D1 (de) | 2008-10-30 |
FR2759474A1 (fr) | 1998-08-14 |
EP0960382B1 (fr) | 2008-09-17 |
US6324475B1 (en) | 2001-11-27 |
JP2001519899A (ja) | 2001-10-23 |
CA2280978A1 (fr) | 1998-08-20 |
FR2759474B1 (fr) | 1999-04-30 |
EP0960382A1 (fr) | 1999-12-01 |
EP0960382B9 (fr) | 2009-03-11 |
RU99119312A (ru) | 2001-08-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP0960382B1 (fr) | Procede pour reconnaitre dans un catalogue preetabli des etoiles detectees par un senseur stellaire | |
EP1298588B1 (fr) | Procédé de traitement d'images pour l'extraction automatique d'éléments sémantiques | |
EP3707676B1 (fr) | Procédé d'estimation de pose d'une caméra dans le référentiel d'une scène tridimensionnelle, dispositif, système de réalite augmentée et programme d'ordinateur associé | |
FR2665597A1 (fr) | Procede et dispositif de localisation en temps reel de contours rectilignes dans une image numerisee, notamment pour la reconnaissance de formes dans un traitement d'analyse de scene. | |
EP1486904A1 (fr) | Génération d'une image type reconstituée à partir d'un ensemble de plusieurs images représentant un même élément | |
FR3094115A1 (fr) | Procede d’identification de bagages | |
EP1327156B1 (fr) | Procede d'individualisation d'un element de circuit integre | |
FR2939583A1 (fr) | Procede d'identification ou d'autorisation, et systeme et module securise asssocies. | |
FR2807852A1 (fr) | Procedes et dispositifs d'indexation et de recherche d'images numeriques prenant en compte la distribution spatiale du contenu des images | |
FR2753820A1 (fr) | Procede de recherche d'echantillons dans des images d'une base de donnees | |
EP3671578A1 (fr) | Procédé d'analyse d'une simulation de l'exécution d'un circuit quantique | |
EP0449353A1 (fr) | Dispositif de traitement de données et procédé pour sélectionner des mots de données contenus dans un dictionnaire | |
EP1532550A2 (fr) | Detection d'une image de reference robuste a de grandes transformations photometriques | |
WO2022008605A1 (fr) | Dispositif électronique et procédé de traitement de données à base de réseaux génératifs inversibles, système électronique de détection et programme d'ordinateur associés | |
EP3828826A1 (fr) | Procede et systeme d'aide a la decision pour la gestion d'une parcelle agricole | |
EP0447306B1 (fr) | Dispositif de reconnaissance de séquences dans un signal multidimensionnel | |
FR3094104A1 (fr) | Procédé et dispositif de détermination de la taille mémoire globale d’une zone mémoire globale allouée aux données d’un réseau de neurones compte tenu de sa topologie | |
FR2704668A1 (fr) | Appareil de poursuite de véhicules routiers à partir d'une plate-forme aérienne. | |
WO2018109030A1 (fr) | Procédé de détermination de l'appartenance d'une donnée à une base de données et produit programme d'ordinateur et support d'informations associés | |
EP1579386A1 (fr) | Procede et dispositif de detection de points d'interet dans une image numerique source, programme d'ordinateur et support de donees correspondants | |
FR3114180A1 (fr) | Système et procédé pour éviter un oubli catastrophique dans un réseau neuronal artificiel | |
EP1211614A1 (fr) | Procédé de recherche dans une structure hiérarchique d'objets | |
WO2021069536A1 (fr) | Procédé de reconnaissance d'objets à représentativité augmentée | |
EP3948664A1 (fr) | Procede et systeme d'identification d'objets a partir d'images labellisees desdits objets | |
EP4078959A1 (fr) | Procede d'encodage d'une image numerique en vue de sa compression |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): CA JP RU US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
ENP | Entry into the national phase |
Ref document number: 2280978 Country of ref document: CA Ref country code: CA Ref document number: 2280978 Kind code of ref document: A Format of ref document f/p: F |
|
WWE | Wipo information: entry into national phase |
Ref document number: 09367428 Country of ref document: US |
|
ENP | Entry into the national phase |
Ref country code: JP Ref document number: 1998 535423 Kind code of ref document: A Format of ref document f/p: F |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1998908183 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 1998908183 Country of ref document: EP |
|
WWG | Wipo information: grant in national office |
Ref document number: 1998908183 Country of ref document: EP |