CN102930244A - Pixel distinguishbility-based natural guide post and method for determining landing area - Google Patents

Pixel distinguishbility-based natural guide post and method for determining landing area Download PDF

Info

Publication number
CN102930244A
CN102930244A CN2012103526113A CN201210352611A CN102930244A CN 102930244 A CN102930244 A CN 102930244A CN 2012103526113 A CN2012103526113 A CN 2012103526113A CN 201210352611 A CN201210352611 A CN 201210352611A CN 102930244 A CN102930244 A CN 102930244A
Authority
CN
China
Prior art keywords
pixel
discrimination
circle
touchdown area
distinguishbility
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
CN2012103526113A
Other languages
Chinese (zh)
Other versions
CN102930244B (en
Inventor
裴明涛
王亚菲
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.)
Beijing Institute of Technology BIT
Original Assignee
Beijing Institute of Technology BIT
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 Beijing Institute of Technology BIT filed Critical Beijing Institute of Technology BIT
Priority to CN201210352611.3A priority Critical patent/CN102930244B/en
Publication of CN102930244A publication Critical patent/CN102930244A/en
Application granted granted Critical
Publication of CN102930244B publication Critical patent/CN102930244B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention relates to a pixel distinguishbility-based natural guide post and a method for determining a landing area, and belongs to the field of deep space exploration. According to the method, on the basis of the characteristic that the distinguishbility of pixel points in a flat area in an image and pixel points in an area which is adjacent to the flat area is low and the distinguishbility of pixel points in areas of meteor craters, rocks and the like and pixel points in areas which are adjacent to the areas is high, pixel points of which the distinguishbility is lower than a preset threshold value are used as candidate points of the landing area, pixel points of which the distinguishbility is higher than the preset threshold value are used as candidate points of the natural guide post, and a set of the candidate points of the natural guide post within a preset radius is used as the natural guide post by calculating the distinguishbility of each pixel point in the image; and morphological expansion is performed on the candidate points of the natural guide post and pixels at the image boundary, the candidate point of the landing area which is covered by expansion operation finally is used as the center of a circle, and distance between the center of the circle and the candidate point, which is close to the center of the circle, of the natural guide post is used as radius to obtain the maximum landing circle. By the method, the landing area can be determined quickly and reliably.

Description

But a kind of natural landmark and touchdown area based on the pixel discrimination determined method
Technical field
The invention belongs to the survey of deep space neighborhood, relate to the correlation technique that image is processed.
Background technology
Natural landmark detects and touchdown area determines it is major issue in the survey of deep space field.In the survey of deep space process, because lander is very remote apart from the earth, the signal transmission has larger time-delay, but therefore needs the attitude of determining self that lander can be autonomous and position and the autonomous touchdown area that finds.
Natural landmark refers in the environment existing, and unartificial setting, can be in order to references object or the pattern that identifies varying environment.Attitude and the position of self determined in the natural landmark detection for lander, carrying out independent navigation has important effect.Existing natural landmark detection method be mostly for planetary surface various landform (such as meteorite crater, rock etc., image appearance features as shown in Figure 1) is carried out natural landmark and is detected.But these class methods are subject to the impact of illumination and concrete landform, can't obtain good result.
And touchdown area determines it also is very important for detector automatically.In the lander landing period, since very remote apart from the earth, rely on ground communications can not satisfy the demand that actual time safety lands, therefore need detector to have the ability of stronger Autonomous determination touchdown area.Existing independent landing area determination method mainly is that the barriers such as rock are determined touchdown area by avoiding obstacles by the meteorite crater of detection planetary surface.Also the textural characteristics of with good grounds planetary surface carries out definite method of touchdown area.The problem of carrying out the definite method of touchdown area by detecting barrier is that the detection of barrier is difficult to accomplish accurately and robust, affected by illumination condition larger.Existing texture analysis method is to be undertaken by the features such as contrast in the statistical regions, also is subject to the impact of the conditions such as illumination.
We make discovery from observation, the barrier of planetary surface, no matter be meteorite crater (Fig. 1 (a)), rock (Fig. 1 (b)), perhaps crackle (Fig. 1 (c)), all have following characteristics: the point on these obstacles borders is (such as 1 among Fig. 1,2,3,4), very high with the degree distinguished of other points in its neighborhood, both be easy to these points and other point are on every side made a distinction.And flat site in point (such as 5,6 among Fig. 1,7), to its around point more similar, substantially can't with make a distinction on every side.We are based on this observation, carry out but natural landmark detected to determine to be unified under the discrimination framework of pixel with touchdown area.
Summary of the invention
Determine the deficiency that method exists with touchdown area and propose that a kind of natural landmark based on the pixel discrimination detects and touchdown area is determined method but the objective of the invention is to detect for above-mentioned natural landmark.
Main contents of the present invention are: the employing video camera obtains the image information below the lander in lander decline process, in the image that each frame gets access to, but calculate the discrimination of each pixel, both the degree distinguished of the pixel in this pixel and its neighborhood.Setting threshold, but discrimination is higher than the pixel of given threshold value as the natural landmark candidate point, with the set of the natural landmark candidate point in the certain radius as natural landmark; But but discrimination is lower than the pixel of given threshold value as the candidate point in safe landing zone, by the pixel on natural landmark candidate point and the image boundary is carried out morphological dilation, to be inflated at last touchdown area candidate point that operation covers as the center of circle, take this center of circle to apart from the distance of its nearest natural landmark candidate point as radius, obtain the maximum circle that can land.
The objective of the invention is to be achieved through the following technical solutions.
But a kind of natural landmark based on the pixel discrimination detects and touchdown area is determined method, and the specific implementation step is as follows:
Step 1: Image Acquisition
Video camera is installed in the below of lander, the image on acquisition ground that can be continuous in lander decline process.
Step 2: but the discrimination of calculating pixel point
If s (p, p ') be pixel p and pixel p ' similarity degree, establish p (s (p, p ')) expression pixel p and pixel p ' similar probability, but the discrimination of definition pixel is this pixel and the entropy of the interior similar probability distribution of pixel of its neighborhood, both
distinguishbility p = - Σ p ′ ∈ N p p ( s ( p , p ′ ) ) log p ( s ( p , p ′ ) ) - - - ( 1 )
N wherein pNeighborhood for pixel p.Entropy is larger, but then discrimination is lower.
Step 3: but but determine touchdown area based on the discrimination of pixel
Barrier region is all regarded in zone but discrimination is high and that but the medium pixel of discrimination covers as, but but touchdown area is regarded in the zone that the low pixel of discrimination covers as.But also regard the point on the image boundary as pixel that discrimination is high, both also regarded barrier zone as.The pixel of barrier region is carried out morphological dilations at every turn, pixel around the barrier also is set to barrier region, along with the carrying out of expanding, but touchdown area is more and more less, but but to be inflated at last the touchdown area that operation covers be exactly the center of circle of maximum touchdown area.But maximum can land radius of a circle can by calculate with the nearest expansion of this point before the height discrimination or the distance of the pixel of medium discrimination obtain
Step 4: the natural landmark based on high discrimination pixel detects
But the pixel with height discrimination can separate with the pixel region around it, can be used as natural landmark.But we with the set of the pixel with height discrimination in the certain radius as natural landmark.
Advantage of the present invention
The present invention and other natural landmark and touchdown area determine that method compares, and have the advantage of the following aspects:
(1) with meteorite crater, but but the detection of the detection of the various natural landmark such as rock and touchdown area is unified under the framework of discrimination processes, can obtain simultaneously natural landmark and touchdown area.
(2) obtain maximum can the land round center of circle and radius by non-natural landmark candidate point being carried out morphologic expansive working, can fast and reliable determine to manage to make do the zone, land.
Description of drawings
The various landform of Fig. 1 planetary surface;
But Fig. 2 determines method flow diagram based on natural landmark and the touchdown area of pixel discrimination;
The similar probability distribution graph of pixel in Fig. 3 pixel and its neighborhood;
Fig. 4 maximum can be landed to justify and be determined the method schematic diagram.
Embodiment
Below in conjunction with drawings and Examples the present invention is elaborated.
But the process flow diagram that natural landmark detection and touchdown area based on the pixel discrimination that the present invention proposes are determined method is as shown in Figure 2, and the implementation step is as follows:
Step 1: Image Acquisition
Video camera is installed in the below of lander, the image on acquisition ground that can be continuous in lander decline process.
Step 2: but the discrimination of calculating pixel point
If a pixel in the p presentation video, its coordinate is (px, py), and its pixel value is I (px, py), but then the discrimination of pixel p is the degree distinguished of other pixels in p and its neighborhood.If the neighborhood of pixel p is N p, N pBe a rectangle centered by pixel p, its length and width can be set according to actual conditions, if its length and width are set as width and the height of image, but then are the discriminations that carries out this pixel in whole sub-picture, and neighborhood N is set pLength and width be respectively picture traverse and the height 1/3rd.Definition s (p, p ') be pixel p and pixel p ' similarity degree, herein, we adopt SSD to calculate the similarity degree of two pixels, and the SSD value is less, show that two pixels are more similar, the Size of Neighborhood that adopts when SSD is calculated in design is (2n+1) * (2m+1), then
s ( p , p ′ ) = Σ i = - n n Σ j = - m m [ I ( px + i , py + j ) - I ( p ′ x + i , p ′ y + j ) ] 2 - - - ( 2 )
Wherein (px, py) is the coordinate of pixel p in image, and (p ' x, p ' are the coordinate of pixel p ' in image y).If p (s (p, p ')) expression pixel p and pixel p ' similar probability, then
p ( s ( p , p ′ ) ) = exp { - s ( p , p ′ ) } Σ p ′ ∈ N p exp { - s ( p , p ′ ) } - - - ( 3 )
But the discrimination of definition pixel be this pixel and the entropy of the interior similar probability distribution of pixel of its neighborhood, both
distinguishbility p = - Σ p ′ ∈ N p p ( s ( p , p ′ ) ) log p ( s ( p , p ′ ) ) - - - ( 4 )
For the pixel of mark among Fig. 1 calculate to its neighborhood in the similar probability distribution of pixel, as shown in Figure 3, can find out, be distributed with three kinds of situations:
1. spot distribution, shown in Fig. 3 (1-2), these pixels generally are the angle points in the image, the entropy of its similar probability distribution is less, but discrimination is high.
2. wire distributes, and shown in Fig. 3 (3-4), these pixels generally are the edges in the image, and the entropy of its similar probability distribution is placed in the middle, but discrimination is medium.
3. planar distribution, shown in Fig. 3 (5-7), these pixels generally belong to flat site, and the entropy of its similar probability distribution is larger, but discrimination is low.
Step 3: but but determine touchdown area based on the discrimination of pixel
Barrier region is all regarded in the zone that but we are high with discrimination and but the medium pixel of discrimination covers as, but but touchdown area is regarded in the zone that the low pixel of discrimination covers as.But touchdown area should satisfy following two requirements, the one, and area is enough large, so that lander lands; The 2nd, enough far away apart from barrier, to guarantee the safety of landing mission.Because the zone of image boundary outside is zone of ignorance, therefore desirable touchdown area should the range image border also enough far.Therefore but we also are set to height with the discrimination of the point on the image boundary, but both borderline point was also regarded the pixel that discrimination is high as.As shown in Figure 4, the some barrier region of redness among the figure, borderline point is also thought barrier region.But green portion is the low pixel of discrimination, can be used as touchdown area, and our target is to find the circle of a maximum, but so that this circle only comprises green area as touchdown area.Specific practice is the pixel of barrier region to be carried out morphological dilations at every turn, pixel around the barrier also is set to barrier region, such as Fig. 4 (b), (c), (d) shown in, can see, along with the carrying out of expanding, green area is more and more less, but the green area that reddens at last is exactly the center of circle of maximum touchdown area.Idiographic flow is:
1) successively the pixel in the barrier region (but pixel of height discrimination and medium discrimination) point is carried out morphological dilations, a pixel expands at every turn; Both the zone with a pixel around the barrier region also was set to barrier region.
2) but judge the number of the pixel that discrimination is low, if number greater than 1, then turns the first step; If number equals 1, then turned for the 3rd step, if number is 0, then turned for the 4th step.
3) if but the number of the low pixel of discrimination equals 1, then this pixel is the maximum round center of circle of can landing, but maximum can land radius of a circle can by calculate with the nearest expansion of this point before the height discrimination or the distance of the pixel of medium discrimination obtain, shown in Fig. 4 (d).
4) if but the number of the low pixel of discrimination equals 0, then choose before this time expansion, but but the low pixel of capped discrimination is as the center of circle of touchdown area, if any a plurality of such pixels, but then get the center of circle near a conduct touchdown area of picture centre, but radius can by calculate with the nearest expansion of this point before the height discrimination or the distance of the pixel of medium discrimination obtain.
Step 4: the natural landmark based on high discrimination pixel detects
Natural landmark refers in the environment existing, and unartificial setting, can be in order to references object or the pattern that identifies varying environment.Definition by natural landmark can find, but the pixel with height discrimination meets the definition of natural landmark, but the pixel that had both had a height discrimination can separate with the pixel region around it, can be used as natural landmark.But we with the set of the pixel with height discrimination in the certain radius as natural landmark.Because but the discrimination of these points is very high, therefore in follow-up image, can more guaranteedly trace into these pixels.

Claims (2)

1. but natural landmark and the touchdown area based on the pixel discrimination determined method, it is characterized in that: but a kind of natural landmark based on the pixel discrimination detects and touchdown area is determined method, and the specific implementation step is as follows:
Step 1: Image Acquisition
Video camera is installed in the below of lander, the image on acquisition ground that can be continuous in lander decline process;
Step 2: but the discrimination of the pixel in the image that calculation procedure one is obtained
If s (p, p ') be pixel p and pixel p ' similarity degree, establish p (s (p, p ')) expression pixel p and pixel p ' similar probability, but the discrimination of definition pixel is this pixel and the entropy of the interior similar probability distribution of pixel of its neighborhood, both
distinguishbility p = - Σ p ′ ∈ N p p ( s ( p , p ′ ) ) log p ( s ( p , p ′ )
N wherein pNeighborhood for pixel p; Entropy is larger, but then discrimination is lower;
Step 3: but but the discrimination of the pixel that calculates based on step 2 is determined touchdown area
Barrier region is all regarded in zone but discrimination is high and that but the medium pixel of discrimination covers as, but but touchdown area is regarded in the zone that the low pixel of discrimination covers as; But also regard the point on the image boundary as pixel that discrimination is high, both also regarded barrier region as; The pixel of barrier region is carried out morphological dilations at every turn, pixel around the barrier also is set to barrier region, along with the carrying out of expanding, but touchdown area is more and more less, but but to be inflated at last the touchdown area that operation covers be exactly the center of circle of maximum touchdown area; But maximum can land radius of a circle can by calculate with the nearest expansion in the center of circle before the height discrimination or the distance of the pixel of medium discrimination obtain;
Step 4: carry out the natural landmark detection based on the high discrimination pixel that step 2 obtains
But the pixel with height discrimination can separate with the pixel region around it, meets the requirement as road sign, can be used as natural landmark; But we are natural landmark with the set of pixels cooperation with height discrimination in the certain radius.
2. but a kind of natural landmark and touchdown area based on the pixel discrimination as claimed in claim 1 determined method, it is characterized in that: can the land round center of circle and radius determining step of the maximum in the described step 2 is:
1) but be that the pixel of height discrimination and medium discrimination carries out morphological dilations to the pixel in the barrier region successively, a pixel expands at every turn; Both the zone with a pixel around the barrier region also was set to barrier region;
2) but judge the number of the pixel that discrimination is low, if number greater than 1, then turns the first step; If number equals 1, then turned for the 3rd step, if number is 0, then turned for the 4th step;
3) if but the number of the low pixel of discrimination equals 1, then this pixel is the maximum round center of circle of can landing, but maximum can be landed radius of a circle can by calculate with the nearest expansion of this point before the height discrimination or the distance of the pixel of medium discrimination obtain;
4) if but the number of the low pixel of discrimination equals 0, then choose before this time expansion, but but the low pixel of capped discrimination is as the center of circle of touchdown area, if any a plurality of such pixels, but then get the center of circle near a conduct touchdown area of picture centre, but radius can by calculate with the nearest expansion of this point before the height discrimination or the distance of the pixel of medium discrimination obtain.
CN201210352611.3A 2012-09-20 2012-09-20 A kind of natural landmark based on pixel may differentiate and touchdown area defining method Expired - Fee Related CN102930244B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210352611.3A CN102930244B (en) 2012-09-20 2012-09-20 A kind of natural landmark based on pixel may differentiate and touchdown area defining method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210352611.3A CN102930244B (en) 2012-09-20 2012-09-20 A kind of natural landmark based on pixel may differentiate and touchdown area defining method

Publications (2)

Publication Number Publication Date
CN102930244A true CN102930244A (en) 2013-02-13
CN102930244B CN102930244B (en) 2016-04-06

Family

ID=47645041

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210352611.3A Expired - Fee Related CN102930244B (en) 2012-09-20 2012-09-20 A kind of natural landmark based on pixel may differentiate and touchdown area defining method

Country Status (1)

Country Link
CN (1) CN102930244B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101539422A (en) * 2009-04-22 2009-09-23 北京航空航天大学 Monocular vision real time distance measure method
CN102173313A (en) * 2010-12-24 2011-09-07 北京控制工程研究所 Soft landing relay obstacle avoiding method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101539422A (en) * 2009-04-22 2009-09-23 北京航空航天大学 Monocular vision real time distance measure method
CN102173313A (en) * 2010-12-24 2011-09-07 北京控制工程研究所 Soft landing relay obstacle avoiding method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
邵巍等: "一种基于形态学的探测器安全着陆点选择方法", 《宇航学报》, vol. 29, no. 2, 31 March 2008 (2008-03-31), pages 659 - 663 *

Also Published As

Publication number Publication date
CN102930244B (en) 2016-04-06

Similar Documents

Publication Publication Date Title
CN103714538B (en) road edge detection method, device and vehicle
CN101661497B (en) Remote sensing land use change detection method and system thereof
CN102768022B (en) Tunnel surrounding rock deformation detection method adopting digital camera technique
CN104484577B (en) The detection method of banding buried target in a kind of mountain region based on ridge energy correction
CN102542551B (en) Automatic change detection technology for floating ice at edges of polar ice sheets
CN106842231A (en) A kind of road edge identification and tracking
CN102081801A (en) Multi-feature adaptive fused ship tracking and track detecting method
CN105158762A (en) Identifying and tracking convective weather cells
CN104700399A (en) Method for demarcating large-deformation landslide displacement field based on high-resolution remote sensing image
CN104536009A (en) Laser infrared composite ground building recognition and navigation method
CN104637073A (en) Zonal underground structure detection method based on sun shade compensation
US20220363263A1 (en) Automated bump and/or depression detection in a roadway
US9239960B2 (en) Three-dimensional object detection device
CN115774444B (en) Path planning optimization method based on sparse navigation map
CN102063707B (en) Mean shift based grey relation infrared imaging target segmentation method
CN109241855B (en) Intelligent vehicle travelable area detection method based on stereoscopic vision
CN105141887A (en) Submarine cable area video alarming method based on thermal imaging
Lee et al. Creation of river terrain data using region growing method based on point cloud data from UAV photography
CN103456027B (en) Time sensitivity target detection positioning method under airport space relation constraint
CN113593040B (en) Method and system for identifying abnormal vegetation growth diseases of dike revetment
CN109635387B (en) Method for obtaining deformation law of mining slope body in gully development mining area
UCHIDA et al. A new method for assessing deep catastrophic landslide susceptibility
CN102930244B (en) A kind of natural landmark based on pixel may differentiate and touchdown area defining method
Wang et al. Building heights estimation using ZY3 data—A case study of Shanghai, China
CN111721302A (en) Method for recognizing and sensing complex terrain features on surface of irregular asteroid

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: 20160406

Termination date: 20160920