CN102254147A - Method for identifying long-distance space motion target based on stellar map matching - Google Patents
Method for identifying long-distance space motion target based on stellar map matching Download PDFInfo
- Publication number
- CN102254147A CN102254147A CN2011100958248A CN201110095824A CN102254147A CN 102254147 A CN102254147 A CN 102254147A CN 2011100958248 A CN2011100958248 A CN 2011100958248A CN 201110095824 A CN201110095824 A CN 201110095824A CN 102254147 A CN102254147 A CN 102254147A
- Authority
- CN
- China
- Prior art keywords
- star
- picture point
- visible light
- light camera
- image points
- 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
Abstract
The invention provides a method for identifying a long-distance space motion target based on stellar map matching. The method comprises the following steps: extracting coordinates of image points of all original images photographed by a visible light camera of a miniature space detector; composing the arranged image points into triangles in a combination mode; reading a star catalogue in the memory of the visible light camera of the miniature space detector, and identifying all triangles formed by the image points by adopting a triangle identification algorithm; and deleting all the identified image points, and remaining the unidentified image points which are image points of the motion target. In the method, the characteristics of relative motion of the visible light camera of the miniature space detector and the motion target is not required to know in advance, the image points of a fixed star are deleted by adopting a star catalogue identification method, and the star catalogue contains all fixed stars in the starry sky, so that the image points of the target can be effectively acquired by adopting the method, and the star catalogue identification method has the advantage of higher interference resistance.
Description
(1) technical field
The present invention relates to spationautics, is exactly a kind of remote space movement target recognition methods based on the star chart coupling specifically.
(2) background technology
Minitype spacecraft mainly is that the moving target in the space (including the aircraft of prior imformation and no prior imformation) is surveyed imaging and caught.For adapting to microminiaturized requirement, the detectivity of the visible light camera of miniature space probe is generally relatively poor, and detection range, investigative range, detection accuracy are very limited, and computing power is limited.Under space environment, the relative motion between starry sky background, moving target and the detector three makes that the visible light camera of miniature space probe is relatively more difficult to the identification of moving target.At present, the main flow algorithm of moving target mainly contains frame-to-frame differences point-score, the method for difference based on edge extracting, image registration method such as combine with method of difference based on edge extracting in the movement background.The frame-to-frame differences point-score is that two two field pictures directly subtract each other before and after utilizing, and generally is that the gray-scale value of respective pixel subtracts each other one by one.This method is often used in Flame Image Process, and particularly effect is better under the motionless target travel situation of background.Utilize frame difference method with moving target object and background separation, to obtain the positional information of moving target.But because aircraft aloft can not be static definitely relatively, therefore same fixed star be unwilling to overlap in the image that the visible light camera of the miniature space probe of two continuous frames is taken fully.Therefore, the application of frame-to-frame differences point-score is restricted.Based on the method for difference of edge extracting is Edge extraction to be combined with the difference of image and a kind of algorithm of forming, has very strong practical value.At first, select suitable threshold, what produce binaryzation carries the edge graph picture; Secondly, same adjacent two two field pictures are implemented calculus of differences, form binary map, thereby generated a frame difference binary map; Once more,, generate the edge image of target, strengthened the movement edge of back piece image in the moving region like this, thereby identified moving target the binary map of edge extracting generation and the binary map logical multiply of difference generation.Because target speed is fast, the influence of smear is big (gray scale is inhomogeneous), makes to be difficult to identify moving target after the logical multiply.Image registration is that adjacent two two field pictures of miniature space probe collection are handled with the method for difference method of combining based on edge extracting, at first this two two field picture is carried out registration, carry out calculus of differences then, differentiated image is implemented binary conversion treatment, generate a frame difference binary map; Secondly, the back two field picture in two two field pictures is carried out edge extracting and binary conversion treatment, produce the binary map of a frame object edge and background edge coexistence; At last, two frame binary map are implemented the logical multiply operation, obtain the edge image of moving target.
The method that the difference algorithm that combines with edge extracting based on Fourier-Mellin image registration extracts moving target is that the picture amplitude spectrum of earlier the visible light camera of miniature space probe being taken is carried out logarithm-pole-change, obtains the anglec of rotation and the scale factor of image subject to registration at transformation space application phase correlation method; Obtain the pairing peak of Fourier inverse transformation by the cross-power spectrum that calculates two regular width of cloth images of process and try to achieve their relative translation, thereby respectively front cross frame image and back two two field pictures are carried out registration, two two field pictures behind the registration are carried out difference, obtain a difference image.With this difference image binaryzation, the back image with this difference image and process edge extracting carries out logical multiply at last, identifies moving target thereby detect.These two kinds of methods can be dispeled background very preferably, but still have the minority background dot not dispel, main cause causes because of registration accuracy is low, because target travel figure speed produces smear soon, make the object edge that extracts produce distortion, can minimal effect identification of targets and location.Except said method, a lot of scholars carry out the algorithm that layering is discerned to fuzzy technology and Application of Neural Network in template matches, these methods are to the recognition efficiency height of moving target, but they will be trained earlier the template under the various situations of moving target, scan in image with the template of moving target then and mate, so calculated amount is big.
The background more complicated of minitype spacecraft environment of living in, adding between complex background, moving target and the minitype spacecraft three has relative motion, makes that the visible light camera of miniature space probe is relatively more difficult to the identification of moving target.Therefore, how the moving target in the recognition image is the primary problem that solves in the remote Technique in Rendezvous and Docking research.Consider the characteristic of relative motion between complex background and the moving target, the registration difference algorithm of current multiple classics is difficult to adapt to the moving target identification of space background.The characteristics of the visible light camera photographic images of miniature space probe: image background mainly is made up of fixed star, can not be linked to be line and other any figures between the fixed star, and this makes method for registering commonly used be difficult to find the registration foundation; The target that is taken is little in imaging, be mingled between the fixed star, therefore in the image that the visible light camera of miniature space probe is taken, there is not obvious characteristics, if the method by the long-pending size of imaging surface is sought moving target from the image that the visible light camera of miniature space probe is taken, then will inevitably lose moving target, therefore, press for seek discrimination height, calculated amount little, can practical method at a distance moving target searched for, discern and follow the tracks of.
(3) summary of the invention
The object of the present invention is to provide a kind of remote space movement target recognition methods based on the star chart coupling.
The object of the present invention is achieved like this: a kind of remote space movement target recognition methods based on the star chart coupling, and step is as follows:
Step 1: extract all miniature space probe visible light cameras and photograph the picpointed coordinate of original image, and preserve all picture point energy;
Step 2: all picture points according to energy rank order from big to small;
Step 3: the picture point after the ordering is formed all triangles according to the mode of combination;
Step 4: read the star catalogue in the miniature space probe visible light camera memories;
Step 5: adopt the triangle recognizer that all triangles that picture point constitutes are discerned;
Step 6: reject all and discerned picture point;
Step 7: keep Unidentified picture point, this Unidentified picture point is exactly the moving target picture point.
The present invention has following advantage: no matter how to move between miniature space probe visible light camera and the moving target, geometric relationship between the fixed star asterism is constant, so the present invention needn't know the characteristic of relative motion between miniature space probe visible light camera and the moving target in advance.Owing to adopt method for recognising star map to reject the fixed star picture point, comprised all fixed stars in the starry sky in the star catalogue, so the present invention can obtain the target picture point efficiently.Because method for recognising star map has very high anti-interference, so the present invention photographs the complexity and the insensitive for noise of original image background to miniature space probe visible light camera.
(4) description of drawings
Fig. 1 for image registration with based on the method for difference block scheme of edge extracting;
Fig. 2 is the visible light camera principle of work synoptic diagram of miniature space probe;
Fig. 3 is the original image that the visible light camera of miniature space probe is taken;
Fig. 4 is the Target Recognition process flow diagram flow chart of miniature space probe;
Fig. 5 is a semi-physical simulation experimental system block scheme;
Fig. 6 is the remote space movement target recognition methods embodiment process flow diagram that the present invention is based on the star chart coupling.
(5) embodiment
The invention will be further described for example below in conjunction with accompanying drawing.
Embodiment 1: in conjunction with Fig. 6, and a kind of remote space movement target recognition methods of the present invention based on the star chart coupling, step is as follows:
Step 1: extract all miniature space probe visible light cameras and photograph the picpointed coordinate of original image, and preserve all picture point energy;
Step 2: all picture points according to energy rank order from big to small;
Step 3: the picture point after the ordering is formed all triangles according to the mode of combination;
Step 4: read the star catalogue in the miniature space probe visible light camera memories;
Step 5: adopt the triangle recognizer that all triangles that picture point constitutes are discerned;
Step 6: reject all and discerned picture point;
Step 7: keep Unidentified picture point, this Unidentified picture point is exactly the moving target picture point.
Embodiment 2: in conjunction with Fig. 2-Fig. 4, the visible light camera principle of work of miniature space probe of the present invention as shown in Figure 2, picture point in the image of its shooting almost is fixed star picture point (as Fig. 3) entirely, can not be linked to be the figure of line and other any routines between the fixed star picture point, this makes method for registering commonly used be difficult to find the registration foundation.Star catalogue has comprised all fixed stars in the celestial sphere, so can utilize star Pattern Recognition Algorithm all star pictures in the image of the visible light camera shooting of miniature space probe are found out corresponding fixed star from star catalogue, pick out the star picture of having discerned, remaining is exactly the moving target picture point.The original image that photographs according to the visible light camera of miniature space probe, therefrom extract the coordinate (these picture points comprise fixed star imaging picture point and target imaging picture point) of all picture points, (the most frequently used feature has magnitude to utilize these picture points to constitute certain feature, star is to angular distance, subtended angle, spectral signature, by geometric configuration of the constellation that a plurality of star picture constituted etc.), the geometric configuration that constitutes with star catalogue in the visible light camera memories that is saved in miniature space probe compares, comparison procedure adopts typical star Pattern Recognition Algorithm to carry out, star Pattern Recognition Algorithm is chosen fixed star in the star catalogue as nautical star, as the observation star, and discern these all observation stars with all picpointed coordinates.Because star catalogue comprised all fixed star information in the celestial sphere, so the picture point of moving target in the original image that the visible light camera of miniature space probe photographs can not be found out corresponding relation in star catalogue.In the picture point of all extractions, pick out the picture point discerned (picture point that can discern be fixed star entirely in the original image that the visible light camera of miniature space probe photographs picture point), remaining is exactly the picture point of moving target.
Detailed process is as follows: at first extract the picpointed coordinate that all miniature space probe visible light cameras photograph original image, and preserve all picture point energy, for the ease of identification, all picture points according to energy rank order from big to small, are formed all triangles to the picture point after the ordering according to the mode of combination; Next reads the star catalogue in the miniature space probe visible light camera memories, adopts the triangle recognizer that all triangles that picture point constitutes are discerned; Reject all once more and discerned picture point, keep Unidentified picture point simultaneously, this Unidentified picture point is exactly the moving target picture point.
Embodiment 3: main performance index of the present invention: the main inspection surface of moving target is about 2m * 2m.At 15km～30km remote section provides illumination by solar radiation and earth light, and the magnitude of moving target is equivalent to 2 stars such as grade～6 stars such as grade.The visible light camera resolution of the miniature space probe of photographic subjects is 1024 * 1024, field angle is 14 ° * 14 °, detectable 6 grades are more than the star, being imaged as of moving target is defocused to a plurality of pixels specially, the visible light camera of miniature space probe has certain disturbance, absolute orientation or during to inertial orientation, 0.1 ° of attitude error, 0.05 °/s of attitude stability; To the moving target orientation time, maximum 0.5 ° of attitude error, 0.1 °/s of attitude stability.The magazine navigation star database of the visible light of miniature space probe comes from the SKY2000 star catalogue, but has used the apparent magnitude at all fixed stars more than the star such as 6.0 grades, and one has 5062.Respectively the magnitude of moving target is equivalent to stars such as 2 stars such as grade, stars such as 2.5, stars such as 3, stars such as 3.5, stars such as 4, stars such as 4.5, stars such as 5, stars such as 5.5 and 6.
Embodiment 4: the semi-physical simulation experimental provision comprises: the track generator, and star field simulation system (comprising many stars analog hardware and software), the visible light camera of miniature space probe, terminal shows computing machine.The user imports the relative motion rule of miniature space probe and moving target, calculate relative attitude and the position of moving target under inertial coordinates system under the visible light camera inertial coordinates system of miniature space probe according to the characteristics of motion, and this attitude and moving target are sent to the star field simulation system in the position under the inertial coordinates system by the RS422 interface, the star field simulation system searches out all fixed stars in the visual field according to this attitude, and calculate all fixed stars and the moving target picpointed coordinate in the picture plane, and demonstrate in real time according to the position of picpointed coordinate relation and to comprise all picture points image of (comprising fixed star picture point and target picture point), the visible light camera of miniature space probe is taken this image in real time, utilize gravity model appoach from the image of taking, to extract all picpointed coordinates, utilize the star catalogue of preserving that these picpointed coordinates are discerned, recognition result is sent to terminal show computing machine, feasibility for recognition methods in the visible light camera of verifying miniature space probe, the star field simulation system shows computing machine to all fixed star information and moving target information in the visual field by wired terminal that sends to, and terminal shows that computing machine utilizes the recognition result of the visible light camera of the known results of star field simulation system and miniature space probe to compare.If the identification asterisk of the visible light camera of miniature space probe output is identical with the known asterisk of star field simulation system, illustrate that this picture point is exactly a fixed star, if the identification asterisk of the visible light camera of miniature space probe output is-1, illustrate that this picture point is exactly the moving target picture point.
Table 1-table 9 is equivalent to 2 stars such as grade to the magnitude of moving target respectively, 2.5 on star, 3 stars such as grade, 3.5 on star, 4 stars such as grade, 4.5 on star, 5 stars such as grade, 5.5 Deng constantly recognition result of picked at random in star and 6 stars such as grade (if in the table identification asterisk of camera output and star field simulation system to export original asterisk consistent, illustrate that this picture point is a fixed star, for the ease of distinguishing, is the identification asterisk of the output of the target picture point camera after the identification-1, the original asterisk of target picture point of star field simulation system output is 10000000), the time of obtaining the moving target picture point is respectively 152.97 milliseconds, 184.75 millisecond, 166.31 millisecond, 127.83 millisecond, 1.642000 millisecond, 119.65 millisecond, 157.30 milliseconds, 197.63 milliseconds, 191.30 millisecond satisfies the real-time on the engineering.
Terminal showed the result that computing machine receives when the magnitude of table 1 target was equivalent to 2 stars such as grade
Terminal showed the result that computing machine receives when the magnitude of table 2 target was equivalent to 2.5 stars such as grade
Terminal showed the result that computing machine receives when the magnitude of table 3 target was equivalent to 3 stars such as grade
Terminal showed the result that computing machine receives when the magnitude of table 4 target was equivalent to 3.5 stars such as grade
Terminal showed the result that computing machine receives when the magnitude of table 5 target was equivalent to 4 stars such as grade
Terminal showed the result that computing machine receives when the magnitude of table 6 target was equivalent to 4.5 stars such as grade
Terminal showed the result that computing machine receives when the magnitude of table 7 target was equivalent to 5 stars such as grade
Terminal showed the result that computing machine receives when the magnitude of table 8 target was equivalent to 5.5 stars such as grade
Terminal showed the result that computing machine receives when the magnitude of table 9 target was equivalent to 6 stars such as grade
Because the picture point of extracting from the image that the visible light camera of miniature space probe is taken the photograph almost is the fixed star picture point entirely, in order to find out the moving target picture point the image of taking the photograph from the visible light camera of miniature space probe, all picture points in the image that can take the photograph the visible light camera of miniature space probe adopt the principle of importance in star map recognition to discern, and judge according to recognition result which may be the moving target picture point like this.Because the visible light camera of miniature space probe points at random, therefore the sensing along with the visible light camera of miniature space probe changes, it is different that the stellar magnitude that is comprised in the image that the visible light camera of miniature space probe is taken does not wait, fixed star distributes, the time of the Image Acquisition moving target picture point of taking from the visible light camera of miniature space probe is different, but the general 100-200 millisecond of these acquisition times is so can satisfy the requirement of embedded system real-time performance.
Embodiment 5: in conjunction with Fig. 6, a kind of remote space movement target recognition methods of the present invention based on the star chart coupling, specific embodiments is as follows: the visible light camera one of miniature space probe powers on, the visible light camera of miniature space probe independently begins photographic images, extract the picture point module and read this image, and adopt gravity model appoach to obtain all picpointed coordinates in the image, and these picpointed coordinates are sorted from big to small according to energy, these picture points are formed all possible triangle according to the mode of permutation and combination, calculate in twos the star between the picture point to angular distance, calculate between each Atria picture point star simultaneously to the angular distance sum, star between three picture points to the angular distance sum according to from big to small rank order, since star between three picture points to angular distance sum maximum, whether search has the triangle that satisfies condition from star catalogue, if have, illustrate and discern successfully that these three picture points are exactly the picture point of fixed star at the visible light camera of miniature space probe; If the triangle that does not satisfy condition illustrates that it is the moving target picture point that three picture points have a picture point at least, this triangle is broken, behind the new triangle of other picture point reformulation, repeat above-mentioned search procedure.After all triangles have been discerned, to add up all and discern unsuccessful picture point quantity, these discern unsuccessful picture point is exactly the picture point of moving target at the visible light camera of miniature space probe, and preserves this picture point.
Claims (1)
1. remote space movement target recognition methods based on the star chart coupling, it is characterized in that: step is as follows:
Step 1: extract all miniature space probe visible light cameras and photograph the picpointed coordinate of original image, and preserve all picture point energy;
Step 2: all picture points according to energy rank order from big to small;
Step 3: the picture point after the ordering is formed all triangles according to the mode of combination;
Step 4: read the star catalogue in the miniature space probe visible light camera memories;
Step 5: adopt the triangle recognizer that all triangles that picture point constitutes are discerned;
Step 6: reject all and discerned picture point;
Step 7: keep Unidentified picture point, this Unidentified picture point is exactly the moving target picture point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110095824 CN102254147B (en) | 2011-04-18 | 2011-04-18 | Method for identifying long-distance space motion target based on stellar map matching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110095824 CN102254147B (en) | 2011-04-18 | 2011-04-18 | Method for identifying long-distance space motion target based on stellar map matching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102254147A true CN102254147A (en) | 2011-11-23 |
| CN102254147B CN102254147B (en) | 2013-01-23 |
Family
ID=44981403
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110095824 Expired - Fee Related CN102254147B (en) | 2011-04-18 | 2011-04-18 | Method for identifying long-distance space motion target based on stellar map matching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102254147B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102630033A (en) * | 2012-03-31 | 2012-08-08 | 彩虹集团公司 | Method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection |
| CN103996027A (en) * | 2014-05-19 | 2014-08-20 | 上海微小卫星工程中心 | Space-based space target recognizing method |
| CN105182678A (en) * | 2015-07-10 | 2015-12-23 | 中国人民解放军装备学院 | System and method for observing space target based on multiple channel cameras |
| CN107749063A (en) * | 2017-10-17 | 2018-03-02 | 厦门理工学院 | A kind of part method for rapidly positioning based on edge extracting and Fourier transformation |
| CN109284707A (en) * | 2018-09-12 | 2019-01-29 | 首都师范大学 | Moving target detection method and device |
| CN109724591A (en) * | 2019-02-27 | 2019-05-07 | 北京航空航天大学 | It is a kind of based on angular distance screening star sensor artificial star screen out method |
| CN110455281A (en) * | 2019-08-14 | 2019-11-15 | 北京理工大学 | Dark small and weak celestial body optics navigation characteristic Imaging Simulation method |
| CN112162245A (en) * | 2020-09-29 | 2021-01-01 | 中国船舶重工集团公司第七二四研究所 | Radar broadband interference identification method based on time-frequency power spectrum projection |
| CN113936229A (en) * | 2021-10-25 | 2022-01-14 | 中国科学院空间应用工程与技术中心 | Spatial dim small target identification method and device based on frequency domain phase correlation |
| CN114820738A (en) * | 2022-06-30 | 2022-07-29 | 中国人民解放军国防科技大学 | Accurate registration method and device for star atlas, computer equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101441082A (en) * | 2008-12-22 | 2009-05-27 | 中国科学院长春光学精密机械与物理研究所 | Rapid triangle star map recognition method |
| CN101968361A (en) * | 2009-07-28 | 2011-02-09 | 韩春好 | Space absolute orientation technology based on starlight observation |
| WO2011015928A2 (en) * | 2009-08-04 | 2011-02-10 | Vesalis | Image-processing method for correcting a target image in accordance with a reference image, and corresponding image-processing device |
-
2011
- 2011-04-18 CN CN 201110095824 patent/CN102254147B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101441082A (en) * | 2008-12-22 | 2009-05-27 | 中国科学院长春光学精密机械与物理研究所 | Rapid triangle star map recognition method |
| CN101968361A (en) * | 2009-07-28 | 2011-02-09 | 韩春好 | Space absolute orientation technology based on starlight observation |
| WO2011015928A2 (en) * | 2009-08-04 | 2011-02-10 | Vesalis | Image-processing method for correcting a target image in accordance with a reference image, and corresponding image-processing device |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102630033A (en) * | 2012-03-31 | 2012-08-08 | 彩虹集团公司 | Method for converting 2D (Two Dimension) into 3D (Three Dimension) based on dynamic object detection |
| CN103996027A (en) * | 2014-05-19 | 2014-08-20 | 上海微小卫星工程中心 | Space-based space target recognizing method |
| CN103996027B (en) * | 2014-05-19 | 2017-04-12 | 上海微小卫星工程中心 | Space-based space target recognizing method |
| CN105182678A (en) * | 2015-07-10 | 2015-12-23 | 中国人民解放军装备学院 | System and method for observing space target based on multiple channel cameras |
| CN105182678B (en) * | 2015-07-10 | 2018-02-02 | 中国人民解放军装备学院 | A kind of system and method based on multichannel camera observation space target |
| CN107749063B (en) * | 2017-10-17 | 2019-09-24 | 厦门理工学院 | A kind of part method for rapidly positioning based on edge extracting and Fourier transformation |
| CN107749063A (en) * | 2017-10-17 | 2018-03-02 | 厦门理工学院 | A kind of part method for rapidly positioning based on edge extracting and Fourier transformation |
| CN109284707A (en) * | 2018-09-12 | 2019-01-29 | 首都师范大学 | Moving target detection method and device |
| CN109284707B (en) * | 2018-09-12 | 2021-07-20 | 首都师范大学 | Moving target detection method and device |
| CN109724591A (en) * | 2019-02-27 | 2019-05-07 | 北京航空航天大学 | It is a kind of based on angular distance screening star sensor artificial star screen out method |
| CN110455281A (en) * | 2019-08-14 | 2019-11-15 | 北京理工大学 | Dark small and weak celestial body optics navigation characteristic Imaging Simulation method |
| CN112162245A (en) * | 2020-09-29 | 2021-01-01 | 中国船舶重工集团公司第七二四研究所 | Radar broadband interference identification method based on time-frequency power spectrum projection |
| CN113936229A (en) * | 2021-10-25 | 2022-01-14 | 中国科学院空间应用工程与技术中心 | Spatial dim small target identification method and device based on frequency domain phase correlation |
| CN114820738A (en) * | 2022-06-30 | 2022-07-29 | 中国人民解放军国防科技大学 | Accurate registration method and device for star atlas, computer equipment and storage medium |
| CN114820738B (en) * | 2022-06-30 | 2022-09-23 | 中国人民解放军国防科技大学 | Accurate registration method and device for star atlas, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102254147B (en) | 2013-01-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102254147B (en) | Method for identifying long-distance space motion target based on stellar map matching | |
| US10303966B2 (en) | Method and system of image-based change detection | |
| CN106651904B (en) | A kind of more extraterrestrial target method for capturing and tracing of width size range | |
| CN107833250B (en) | Semantic space map construction method and device | |
| CN102128623B (en) | Fast star map identification method of star sensor | |
| CN103426165A (en) | Precise registration method of ground laser-point clouds and unmanned aerial vehicle image reconstruction point clouds | |
| CN106503248A (en) | Ground drawing generating method and map creation device | |
| CN111899345B (en) | Three-dimensional reconstruction method based on 2D visual image | |
| Gupta et al. | Augmented reality system using lidar point cloud data for displaying dimensional information of objects on mobile phones | |
| CN110400374A (en) | The method for building up of panorama point cloud data and establish system | |
| Chen et al. | Camera geolocation from mountain images | |
| Cai et al. | Improving CNN-based planar object detection with geometric prior knowledge | |
| Cao et al. | Detection method based on image enhancement and an improved faster R-CNN for failed satellite components | |
| Koizumi et al. | Development of attitude sensor using deep learning | |
| CN116862832A (en) | Three-dimensional live-action model-based operator positioning method | |
| Konishi et al. | Detection of target persons using deep learning and training data generation for Tsukuba challenge | |
| Yang et al. | Method for building recognition from FLIR images | |
| Ying et al. | Fully Convolutional Networks tor Street Furniture Identification in Panorama Images. | |
| CN111024063A (en) | Star map recognition algorithm under large-mobility condition based on star point re-extraction | |
| CN111080685A (en) | Airplane sheet metal part three-dimensional reconstruction method and system based on multi-view stereoscopic vision | |
| CN114972720B (en) | Autonomous image perception-based high-precision unmanned positioning method | |
| Chen et al. | Remote Sensing Image Registration based on Attention and Residual Network | |
| Qiao et al. | The application of spatial scene matching based on SURF in cruise missile terminal guidance | |
| Armenakis et al. | Feasibility study for pose estimation of small UAS in known 3D environment using geometric hashing | |
| CN107273850B (en) | Autonomous following method based on mobile robot |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130123 |