CN108038876B - Method and device for detecting and estimating cubic component of space target - Google Patents
Method and device for detecting and estimating cubic component of space target Download PDFInfo
- Publication number
- CN108038876B CN108038876B CN201711445395.6A CN201711445395A CN108038876B CN 108038876 B CN108038876 B CN 108038876B CN 201711445395 A CN201711445395 A CN 201711445395A CN 108038876 B CN108038876 B CN 108038876B
- Authority
- CN
- China
- Prior art keywords
- plane
- cube
- planes
- cubic
- estimation
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Navigation (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a device for detecting and estimating a spatial target cube component, wherein the method comprises the following steps: acquiring a cubic component to be estimated; detecting the cube according to the relative spatial position between every two surfaces of the cube; and carrying out space attitude estimation and cube size estimation on the cube according to a robust estimation method. The method utilizes the relative spatial position relation between every two surfaces of the cube to detect and recognize the cube, and estimates the posture and the size of the cube by a robust estimation method, thereby obtaining a complete cube structure and enhancing the expression capability of a geometric model to a target structure. The method can be used for a space target plane structure model, can be popularized and applied to other target plane models with cubic structures, and has wide application.
Description
Technical Field
The invention relates to the field of digital geometric processing, in particular to a method and a device for detecting and estimating a spatial target cube component.
Background
The three-dimensional sensor scans or utilizes a target sequence image to carry out three-dimensional reconstruction, a space three-dimensional point cloud model of a target can be obtained, and tasks such as space target tracking identification, autonomous intersection and butt joint, on-orbit service and the like are applied. The points in the point cloud model only have spatial position information and lack target structure information, so that specific structural analysis is required in practical application, and conventional processing is to detect and identify geometric primitives such as planes, cylinders, cones and the like. Cubes, a common geometry, exist in large numbers on spatial objects. However, since the cube cannot be described by mathematical expressions like other geometric primitives, it cannot be detected by conventional detection methods (such as three-dimensional Hough transform).
Disclosure of Invention
In view of the above problems, the present invention provides a method and an apparatus for detecting and estimating a spatial target cube component, so as to detect and identify a plane belonging to a cube from a spatial target planar structure, and finally estimate a cube structure, thereby improving the expression capability of a spatial target geometric structure model.
In order to achieve the purpose, the invention adopts the following technical scheme:
in a first aspect, an embodiment of the present invention provides a method for detecting and estimating a spatial target cube component, including:
acquiring a cubic component to be estimated;
detecting the cube according to the relative spatial position between every two surfaces of the cube;
and carrying out space attitude estimation and cube size estimation on the cube according to a robust estimation method.
In one embodiment, the detecting of the cube according to the relative spatial position between two surfaces of the cube comprises: judging the position of a cube plane and determining the cube plane;
the judgment of the cube plane position is a plane pair { Pp,PqThe judgment of the relative spatial position comprises the position relationship of two cubic planes:
the opposite cubic plane discrimination condition: i) plane normal vector parallel, np||nq(ii) a ii) the edge vectors are correspondingly parallel, xp||xq,yp||yqOr xp||yq,yp||xq(ii) a iii) the planes are positioned opposite and of similar dimensions,
adjacent cube face discrimination conditions: i) plane normal vector is vertical: n isp⊥nq(ii) a ii) the edge vectors correspond to parallel or perpendicular, xp||xq,xp⊥yq,yp⊥xq,yp⊥yqOr yp||xq,yp⊥yq,xp⊥xq,xp⊥yq(ii) a iii) the planes are adjacently positioned and have similar dimensions, such that PpX ofpEdge and PqX ofqAdjacent, then need to satisfy
Wherein c is a reference vertex of plane P ═ Π (c, n, x, y), n is a unit normal vector of plane P, and x and y are edge vectors of plane P; subscripts p and q correspond to different planes; if (a, b)<TθThen consider a | | b, if |90 ° - < a, b-<TθConsidering a ^ b, wherein the angle (a, b) is the included angle between a and b; will plane PaTo plane PbProjection gave P'aThen P isaAnd PbHas a cross-over ratio ofS (-) is the planar area; dabIs a plane PaCenter point to plane PbVector D of center pointab=(ca+(xa+ya)/2)-(cb+(xb+yb)/2);
The determination of the cube plane comprises:
b) Set of traversal planesIn all plane pairs { Pp,PqIf { P }p,PqSatisfy the opposite plane relation or the adjacent plane relation, then willTaking the cubic plane as a reference cubic plane group and entering the next step; if not, then,there is no cubic plane present;
c) set of traversal planesAll planes P inmIf, ifPmAndall planes in (1) satisfy the opposite plane relation or the adjacent plane relation, andsatisfying a cubic configuration (one surface with only one opposite face, four adjacent faces in different orientations), P will bemIs contained inNamely, it is
d) If it is notThe number of the middle counter planes is more thanNumber of opposing planes in, orNumber of median opposition planes equal toAnd the number of opposite planes inTotal number of midplane is greater thanTotal number of planes in (1) isReplacement of
e) If it is notIn which there is only one set of opposing planes, orIf the total number of planes is not less than three, the number is considered to beIs an effective cubic plane set, pairPerforming cubic space attitude estimation and cubic size estimation, and obtaining a cubic space attitude estimation and a cubic size estimationIn the middle of removingTraversing the steps of a) to d) untilNo valid cube-plane groups are extracted.
In one embodiment, spatial pose estimation of a cube comprises:
for the input deviceCube plane groupRespectively estimating three edge vectors k belonging to { u, v, w };
estimating the direction of the cubic edge vector according to the cubic plane, wherein the formula is as follows:
whereinIs a cube plane group, k belongs to { u, v, w } corresponding to the cube edge;andare respectively asA plane perpendicular to k and parallel to k; δ (·, ·) is an exemplary function, δ (a, b) being 1, if a | | b, otherwise δ (a, b) being 0. W (-) is a weight function, and for a plane normal vector n, W (n) is defined as the reciprocal of the root mean square of the total distance error from all plane points to the plane; for the plane edge vector iota ∈ { x, y }, the definition of W (iota) is the degree of uniformity of the distribution of the plane points along the iota direction: projecting the plane points to iota, arranging the projected points in sequence and equally dividing the projected points into NbinSegment, then
Wherein s isiIs the proportion of the proxels falling within the ith segment.
In one embodiment, the size estimation of the cube comprises:
for input cube plane groupsAnd the estimated edge vectorEstimating cube in each directionLength of dimension (d);
estimating the length of each side of the cube according to the plane of the cube, wherein the calculation formula is as follows:
whereinEdges of the corresponding cube;is composed ofThe projection points from the top points of all the planes to k;andare two clusters distributed at two ends of the k directionThe projected point of (2) is,for a cluster close to the reference origin o,another cluster away from the reference origin o;is composed ofIs comprised of the projection points of the vertices of all planes perpendicular to k.
In a second aspect, an embodiment of the present invention provides a spatial object cube component detection and estimation apparatus, including:
the acquisition module is used for acquiring a cubic component to be estimated;
the detection module is used for detecting the cube according to the relative spatial position between every two surfaces of the cube;
and the estimation module is used for carrying out space attitude estimation and cube size estimation on the cube according to a robust estimation method.
In one embodiment, the detection module includes:
the judgment submodule is used for judging the position of the plane of the cube;
a determination submodule for determining a cube plane;
the judgment of the cube plane position is a plane pair { Pp,PqThe judgment of the relative spatial position comprises the position relationship of two cubic planes:
the opposite cubic plane discrimination condition: i) plane normal vector parallel, np||nq(ii) a ii) the edge vectors are correspondingly parallel, xp||xq,yp||yqOr xp||yq,yp||xq(ii) a iii) the planes are positioned opposite and of similar dimensions,
adjacent cube face discrimination conditions: i) plane normal vector is vertical: n isp⊥nq(ii) a ii) edge vector correspondenceParallel or perpendicular, xp||xq,xp⊥yq,yp⊥xq,yp⊥yqOr yp||xq,yp⊥yq,xp⊥xq,xp⊥yq(ii) a iii) the planes are adjacently positioned and have similar dimensions, such that PpX ofpEdge and PqX ofqAdjacent, then need to satisfy
Wherein c is a reference vertex of plane P ═ Π (c, n, x, y), n is a unit normal vector of plane P, and x and y are edge vectors of plane P; subscripts p and q correspond to different planes; if (a, b)<TθThen consider a | | b, if |90 ° - < a, b-<TθConsidering a ^ b, wherein the angle (a, b) is the included angle between a and b; will plane PaTo plane PbProjection gave P'aThen P isaAnd PbHas a cross-over ratio ofS (-) is the planar area; dabIs a plane PaCenter point to plane PbVector D of center pointab=(ca+(xa+ya)/2)-(cb+(xb+yb)/2);
The determination of the cube plane comprises:
b) Set of traversal planesIn all plane pairs { Pp,PqIf { P }p,PqSatisfy the opposite plane relation or the adjacent plane relation, then willTaking the cubic plane as a reference cubic plane group and entering the next step; if not, then,there is no cubic plane present;
c) set of traversal planesAll planes P inmIf, ifPmAndall planes in (1) satisfy the opposite plane relation or the adjacent plane relation, andsatisfying a cubic configuration (one surface with only one opposite face, four adjacent faces in different orientations), P will bemIs contained inNamely, it is
d) If it is notThe number of the middle counter planes is more thanNumber of opposing planes in, orNumber of median opposition planes equal toAnd the number of opposite planes inTotal number of midplane is greater thanTotal number of planes in (1) isReplacement of
e) If it is notIn which there is only one set of opposing planes, orIf the total number of planes is not less than three, the number is considered to beIs an effective cubic plane set, pairPerforming cubic space attitude estimation and cubic size estimation, and obtaining a cubic space attitude estimation and a cubic size estimationIn the middle of removingTraversing the steps of a) to d) untilNo valid cube-plane groups are extracted.
In one embodiment, the estimation module is configured to perform spatial pose estimation on a cube, and includes:
estimating the direction of the cubic edge vector according to the cubic plane, wherein the formula is as follows:
whereinIs a cube plane group, k belongs to { u, v, w } corresponding to the cube edge;andare respectively asA plane perpendicular to k and parallel to k; δ (·, ·) is an exemplary function, δ (a, b) being 1, if a | | b, otherwise δ (a, b) being 0. W (-) is a weight function, and for a plane normal vector n, W (n) is defined as the reciprocal of the root mean square of the total distance error from all plane points to the plane; for the plane edge vector iota ∈ { x, y }, the definition of W (iota) is the degree of uniformity of the distribution of the plane points along the iota direction: projecting the plane points to iota, arranging the projected points in sequence and equally dividing the projected points into NbinSegment, then
Wherein s isiIs the proportion of the proxels falling within the ith segment.
In one embodiment, the estimation module is further configured to perform size estimation on the cube, and includes:
for input cube plane groupsAnd the estimated edge vectorEstimating cube in each directionLength of dimension (d);
estimating the length of each side of the cube according to the plane of the cube, wherein the calculation formula is as follows:
whereinEdges of the corresponding cube;is composed ofThe projection points from the top points of all the planes to k;andare two clusters distributed at two ends of the k directionThe projected point of (2) is,for a cluster close to the reference origin o,another cluster away from the reference origin o;is composed ofIs comprised of the projection points of the vertices of all planes perpendicular to k.
The technical scheme provided by the embodiment of the invention has the beneficial effects that at least:
according to the technical scheme, compared with the prior art, the method and the device for detecting and estimating the spatial target cube component are disclosed and provided, the method utilizes the relative spatial position relation between every two surfaces of the cube to detect and recognize the cube, and estimates the posture and the size of the cube through a robust estimation method, so that a complete cube structure is obtained, and the expression capacity of a geometric model on the target structure is enhanced. The method can be used for a space target plane structure model, can be popularized and applied to other target plane models with cubic structures, and has wide application.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart of a method for detecting and estimating a cubic component of a spatial target according to an embodiment of the present invention;
FIG. 2 is a spatial plane structure composition definition provided by an embodiment of the present invention;
FIG. 3 is a spatial cube structure composition definition provided by an embodiment of the present invention;
FIG. 4 is a pseudo code diagram of an algorithm of a cube detection and estimation method according to an embodiment of the present invention;
FIGS. 5A1-5A3, 5B1-5B3 are comparative graphs of cube detection estimates provided by embodiments of the present invention;
fig. 6 is a block diagram of a spatial object cube component detection and estimation apparatus according to an embodiment of the present invention.
Detailed Description
For a better understanding of the technical aspects of the present invention, reference will now be made in detail to the embodiments of the present invention as illustrated in the accompanying drawings.
In the following description, various aspects of the invention will be described, however, it will be apparent to those skilled in the art that the invention may be practiced with only some or all of the structures or processes of the present invention. Specific numbers, configurations and sequences are set forth in order to provide clarity of explanation, but it will be apparent that the invention may be practiced without these specific details. In other instances, well-known features have not been set forth in detail in order not to obscure the invention.
The space cube is composed of six surfaces, the six surfaces meet the visual geometrical relationship constraint that every two surfaces are opposite surfaces or adjacent surfaces, and meanwhile, one surface has only one opposite surface and four adjacent surfaces in different directions. The relative spatial position relationship of the cubes can be used as a standard detection of cube surface detection, so that the cube surface plane can be detected and identified from a plurality of spatial rectangular planes.
Aiming at the detection and identification of the spatial target geometric structure, the embodiment of the invention provides a detection and estimation method of a spatial target cube, which detects the cube plane through the relative spatial position relationship between the cube surfaces, estimates the final cube structure and improves the expression capability of a target geometric structure model. The embodiment of the invention discloses a method for detecting and estimating a cubic component of a space target, which is shown in figure 1 and comprises the following steps:
s101, obtaining a cubic component to be estimated;
s102, detecting the cube according to the relative spatial position between every two surfaces of the cube;
s103, carrying out space attitude estimation and cube size estimation on the cube according to a robust estimation method.
The specific implementation flow of the invention is shown in fig. 1, the composition definition of the spatial plane structure is shown in fig. 2, the composition definition of the spatial cube structure is shown in fig. 3, and the implementation algorithm of the cube detection estimation method is shown in fig. 4.
The object processed in this embodiment is a planar structure of the target, and the planar structure can be obtained by performing planar detection (e.g., three-dimensional Hough transform) on a target three-dimensional point cloud (i.e., a three-dimensional point data set of the outer surface of the target). In the present embodiment, the plane in the input target plane structure is represented by P ═ Π (c, n, x, y), where c is a reference vertex of the plane P, n is a unit normal vector of the plane P, x and y are edge vectors of the plane P, and lengths are | x | ═ c, respectivelyx,|y|=cyTaking the modulus of the vector, and forming a right-hand coordinate system by c-xyn; the cube for detection estimation is denoted as C ═ Ω (C, u, v, w), C is a reference vertex of the cube C, u, v and w are cube C edge vectors, and the lengths are | u | ═ C, respectivelyu,|v|=cv,|w|=cwC-uvw constitutes a right-hand coordinate system.
The specific implementation details of each part are as follows:
cube detection
Plane pair { Pp,PqThe judgment of the relative spatial position comprises the position relationship of two cubic planes:
the opposite cubic plane discrimination condition: i) plane normal vector parallel, np||nq(ii) a ii) the edge vectors are correspondingly parallel, xp||xq,yp||yqOr xp||yq,yp||xq(ii) a iii) the planes are positioned opposite and of similar dimensions,
adjacent cube face discrimination conditions: i) plane normal vector is vertical: n isp⊥nq(ii) a ii) the edge vectors correspond to parallel or perpendicular, xp||xq,xp⊥yq,yp⊥xq,yp⊥yqOr yp||xq,yp⊥yq,xp⊥xq,xp⊥yq(ii) a iii) the planes are adjacently positioned and have similar dimensions, such that PpX ofpEdge and PqX ofqAdjacent, then need to satisfy
Wherein if ≈ (a, b)<TθThen consider a | | b, if |90 ° - < a, b-<TθConsidering a ^ b, wherein the angle (a, b) is the included angle between a and b; will plane PaTo plane PbProjection gave P'aThen P isaAnd PbHas a cross-over ratio ofS (-) is the planar area; dabIs a plane PaCenter point to plane PbVector D of center pointab=(ca+(xa+ya)/2)-(cb+(xb+yb)/2)。
If the plane pair { Pp,PqSatisfying all conditions in a) } Pp,PqThe four are opposite faces in a cube; if { P }p,PqSatisfying all conditions in b) then Pp,PqIs a pair of adjacent faces in the cube.
The determination of the cube plane comprises the following steps:
b) Set of traversal planesIn all plane pairs { Pp,PqIf { P }p,PqSatisfy the opposite plane relation or the adjacent plane relation, then willTaking the cubic plane as a reference cubic plane group and entering the next step; if not, then,there is no cubic plane present;
c) set of traversal planesAll planes P inmIf, ifPmAndin (1)All planes satisfy the opposite plane relation or the adjacent plane relation, andsatisfying a cubic configuration (one surface with only one opposite face, four adjacent faces in different orientations), P will bemIs contained inNamely, it is
d) If it is notThe number of the middle counter planes is more thanNumber of opposing planes in, orNumber of median opposition planes equal toAnd the number of opposite planes inTotal number of midplane is greater thanTotal number of planes in (1) isReplacement of
e) If it is notThere is only one groupOpposite planes, orIf the total number of planes is not less than three, the number is considered to beIs an effective cubic plane set, pairPerforming cube estimation of step two and step three, and obtainingIn the middle of removingTraversing the steps of a) to d) untilNo valid cube-plane groups are extracted.
Cubic spatial attitude estimation
For input cube plane groupsAnd respectively estimating three edge vectors k belonging to { u, v, w }, and comprising the following steps:
according to the relative position relationship between the plane and kThe plane in (1) is divided into two types of planes perpendicular to k and parallel to k:and
estimate k
Where δ (·, ·) is an exemplary function, δ (a, b) ═ 1, if a | | b, else δ (a, b) ═ 0. W (-) is a weight function, and for a plane normal vector n, W (n) is defined as the reciprocal of the root mean square of the total distance error from all plane points to the plane; for the plane edge vector iota ∈ { x, y }, the definition of W (iota) is the degree of uniformity of the distribution of the plane points along the iota direction: projecting the plane points to iota, arranging the projected points in sequence and equally dividing the projected points into NbinSegment, then
Wherein s isiIs the proportion of the proxels falling within the ith segment.
constructing matrix by using estimated cubic unit edge vectorTo pairPerforming SVD decompositionThe sigma is forced to be an identity matrix I, so that orthogonal identity side vectors are obtained
Cube size estimation
For input cube plane groupsAnd the estimated edge vectorEstimating cube in each directionThe length of the dimension is divided into the following steps:
1) will be provided withThe vertexes of all the planes in the image are projected to k to obtain a projection point set
2)Will be distributed at both ends in the k direction, and is denoted asAndwhereinFor a cluster of proxels close to the reference origin o,another cluster of projection points far from the reference origin o;
whereinIs composed ofIs comprised of the projection points of the vertices of all planes perpendicular to k. The length dimension of the cube C in the direction k is
4) The finally estimated edge vector of the cube C ═ Ω (C, u, v, w) is
The reference vertex position is
The comparison graphs of the results of the cube detection estimation of the plane structures of the "GPS" model and the "Helios" model by the method of the embodiment are shown in FIGS. 5A1-5A3 and 5B1-5B 3. Fig. 5a1-5A3 show the plane structure of the "GPS" model before cube detection, the estimation result of the "GPS" plane structure cube detection, and the geometry structure after point cloud removal, respectively, and fig. 5B1-5B3 correspond to the results of the "Helios" model. It is obvious from the comparison of the results, that the method of the embodiment effectively detects and estimates the correct cubic structure from the target plane structure, and improves the expression capability of the model.
The invention fully utilizes the relative spatial position relationship between every two surfaces of the cube, carries out cube detection and identification on the basis of a geometric plane structure model, and estimates the spatial posture and the size of the cube by utilizing the cube plane obtained by detection. After the processing by the method, planes originally belonging to the cube are detected and identified and finally estimated to obtain a corresponding cube structure, so that the expression capacity of the geometric model on the target structure is improved; the method has simple and intuitive principle, small calculation complexity and stable and reliable result. The method can be used for a space target plane structure model, can be popularized and applied to other target plane models with cubic structures, and has wide application.
Based on the same inventive concept, the embodiment of the present invention further provides a device for detecting and estimating a spatial target cube component, and as the principle of the problem solved by the device is similar to that of the method for detecting and estimating a spatial target cube component, the implementation of the device may refer to the implementation of the method, and repeated details are omitted.
An embodiment of the present invention provides a spatial target cube component detection and estimation apparatus, which is shown in fig. 6, and includes:
an obtaining module 61, configured to obtain a cubic component to be estimated;
the detection module 62 is used for detecting the cube according to the relative spatial position between every two surfaces of the cube;
and the estimation module 63 is used for performing space attitude estimation and cube size estimation on the cube according to a robust estimation method.
In one embodiment, the detection module 62 includes:
the discrimination submodule 621 is used for discriminating the position of the cube plane;
a determination sub-module 622 for determination of the cube plane;
the judgment of the cube plane position is a plane pair { Pp,PqThe judgment of the relative spatial position comprises the position relationship of two cubic planes:
the opposite cubic plane discrimination condition: i) plane normal vector parallel, np||nq(ii) a ii) the edge vectors are correspondingly parallel, xp||xq,yp||yqOr xp||yq,yp||xq(ii) a iii) the planes are positioned oppositely and have similar sizes,
Adjacent cube face discrimination conditions: i) plane normal vector is vertical: n isp⊥nq(ii) a ii) the edge vectors correspond to parallel or perpendicular, xp||xq,xp⊥yq,yp⊥xq,yp⊥yqOr yp||xq,yp⊥yq,xp⊥xq,xp⊥yq(ii) a iii) the planes are adjacently positioned and have similar dimensions, such that PpX ofpEdge and PqX ofqAdjacent, then need to satisfy
Wherein c is a reference vertex of plane P ═ Π (c, n, x, y), n is a unit normal vector of plane P, and x and y are edge vectors of plane P; subscripts p and q correspond to different planes; if (a, b)<TθThen consider a | | b, if |90 ° - < a, b-<TθConsidering a ^ b, wherein the angle (a, b) is the included angle between a and b; will plane PaTo plane PbProjection gave P'aThen P isaAnd PbHas a cross-over ratio ofS (-) is the planar area; dabIs a plane PaCenter point to plane PbVector D of center pointab=(ca+(xa+ya)/2)-(cb+(xb+yb)/2);
The determination of the cube plane comprises:
b) Set of traversal planesIn all plane pairs { Pp,PqIf { P }p,PqSatisfy the opposite plane relation or the adjacent plane relation, then willTaking the cubic plane as a reference cubic plane group and entering the next step; if not, then,there is no cubic plane present;
c) set of traversal planesAll planes P inmIf, ifPmAndall planes in (1) satisfy the opposite plane relation or the adjacent plane relation, andsatisfying a cubic configuration (one surface with only one opposite face, four adjacent faces in different orientations), P will bemIs contained inNamely, it is
d) If it is notThe number of the middle counter planes is more thanNumber of opposing planes in, orNumber of median opposition planes equal toAnd the number of opposite planes inTotal number of midplane is greater thanTotal number of planes in (1) isReplacement of
e) If it is notIn which there is only one set of opposing planes, orIf the total number of planes is not less than three, the number is considered to beIs an effective cubic plane set, pairPerforming cubic space attitude estimation and cubic size estimation, and obtaining a cubic space attitude estimation and a cubic size estimationIn the middle of removingTraversing the steps of a) to d) untilNo valid cube-plane groups are extracted.
In one embodiment, the estimation module 63 is configured to perform spatial pose estimation on a cube, and includes:
estimating the direction of the cubic edge vector according to the cubic plane, wherein the formula is as follows:
whereinIs a cube plane group, k belongs to { u, v, w } corresponding to the cube edge;andare respectively asA plane perpendicular to k and parallel to k; δ (·, ·) is an exemplary function, δ (a, b) being 1, if a | | b, otherwise δ (a, b) being 0. W (-) is a weight function, and for a plane normal vector n, W (n) is defined as the reciprocal of the root mean square of the total distance error from all plane points to the plane; determination of the plane edge vector iota ∈ { x, y }, W (iota)Meaning the degree of uniformity of the distribution of the planar points along the iota direction: projecting the plane points to iota, arranging the projected points in sequence and equally dividing the projected points into NbinSegment, then
Wherein s isiIs the proportion of the proxels falling within the ith segment.
In one embodiment, the estimating module 63 is further configured to perform size estimation on the cube, and includes:
for input cube plane groupsAnd the estimated edge vectorEstimating cube in each directionLength of dimension (d);
estimating the length of each side of the cube according to the plane of the cube, wherein the calculation formula is as follows:
whereinEdges of the corresponding cube;is composed ofAll planes inThe vertex of (2) to k;andare two clusters distributed at two ends of the k directionThe projected point of (2) is,for a cluster close to the reference origin o,another cluster away from the reference origin o;is composed ofIs comprised of the projection points of the vertices of all planes perpendicular to k.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described device embodiments are merely illustrative, for example, the division of the unit is only a logical functional division, and there may be other division ways in actual implementation, such as: multiple units or components may be combined, or may be integrated into another system, or some features may be omitted, or not implemented. In addition, the coupling, direct coupling or communication connection between the components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between the devices or units may be electrical, mechanical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed on a plurality of network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, all the functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may be separately used as one unit, or two or more units may be integrated into one unit; the integrated unit can be realized in a form of hardware, or in a form of hardware plus a software functional unit.
Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media capable of storing program codes, such as a removable Memory device, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, and an optical disk.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (6)
1. A method for detecting and estimating a component of a spatial target cube, comprising:
acquiring a cubic component to be estimated;
detecting the cube according to the relative spatial position between every two surfaces of the cube;
performing space attitude estimation and cube size estimation on a cube according to a robust estimation method;
the detection of the cube is carried out according to the relative spatial position between every two surfaces of the cube, and the detection comprises the following steps: judging the position of a cube plane and determining the cube plane;
the judgment of the cube plane position is a plane pair { Pp,PqThe judgment of the relative spatial position comprises the position relationship of two cubic planes:
the opposite cubic plane discrimination condition: i) plane normal vector parallel, np||nq(ii) a ii) the edge vectors are correspondingly parallel, xp||xq,yp||yqOr xp||yq,yp||xq(ii) a iii) the planes are positioned opposite and of similar dimensions,
adjacent cube face discrimination conditions: i) plane normal vector is vertical: n isp⊥nq(ii) a ii) the edge vectors correspond to parallel or perpendicular, xp||xq,xp⊥yq,yp⊥xq,yp⊥yqOr yp||xq,yp⊥yq,xp⊥xq,xp⊥yq(ii) a iii) the planes are adjacently positioned and have similar dimensions, such that PpX ofpEdge and PqX ofqAdjacent, then need to satisfy
Wherein c is a reference vertex of plane P ═ Π (c, n, x, y), n is a unit normal vector of plane P, and x and y are edge vectors of plane P; subscripts p and q correspond to different planes; if (a, b)<TθThen consider a | | b, if |90 ° - < a, b-<TθConsidering a ^ b, wherein the angle (a, b) is the included angle between a and b; will plane PaTo plane PbProjection gave P'aThen P isaAnd PbHas a cross-over ratio ofS (-) is the planar area; dabIs a plane PaCenter point to plane PbVector D of center pointab=(ca+(xa+ya)/2)-(cb+(xb+yb)/2);
The determination of the cube plane comprises:
b) Set of traversal planesIn all plane pairs { Pp,PqIf { P }p,PqWhen the opposite cube face discrimination condition or the adjacent cube face discrimination condition is satisfied, theTaking the cubic plane as a reference cubic plane group and entering the next step; if not, then,there is no cubic plane present;
c) set of traversal planesAll planes P inmIf, ifPmAndall planes in (1) satisfy the opposite plane relation or the adjacent plane relation, andthe cube configuration is satisfied: one surface has only one opposite surface and four adjacent surfaces in different directions, P ismIs contained inNamely, it is
d) If it is notThe number of the middle counter planes is more thanNumber of opposing planes in, orNumber of median opposition planes equal toAnd the number of opposite planes inTotal number of midplane is greater thanTotal number of planes in (1) isReplacement of
e) If it is notIn which at least one set of opposing planes is present, orIf the total number of planes is not less than three, the number is considered to beIs an effective cubic plane set, pairPerforming cubic space attitude estimation and cubic size estimation, and obtaining a cubic space attitude estimation and a cubic size estimationIn the middle of removingTraversing the steps of a) to d) untilNo valid cube-plane groups are extracted.
2. The method of claim 1, wherein performing spatial pose estimation on a cube comprises:
estimating the direction of the cubic edge vector according to the cubic plane, wherein the formula is as follows:
whereinIs a cube plane group, k belongs to { u, v, w } corresponding to the cube edge;andare respectively asA plane perpendicular to k and parallel to k; δ (·, ·) is an exemplary function, δ (a, b) equals 1 if a | | b, otherwise δ (a, b) equals 0; w (-) is a weight function, and for a plane normal vector n, W (n) is defined as the reciprocal of the root mean square of the total distance error from all plane points to the plane; for the plane edge vector iota ∈ { x, y }, the definition of W (iota) is the degree of uniformity of the distribution of the plane points along the iota direction: projecting the plane points to iota, arranging the projected points in sequence and equally dividing the projected points into NbinSegment, then
Wherein s isiIs the proportion of the proxels falling within the ith segment.
3. The method of any of claims 1-2, wherein performing size estimation on the cube comprises:
for input cube plane groupsAnd the estimated edge vectorEstimating cube in each directionLength of dimension (d);
estimating the length of each side of the cube according to the plane of the cube, wherein the calculation formula is as follows:
whereinEdges of the corresponding cube;is composed ofThe projection points from the top points of all the planes to k;andare two clusters distributed at two ends of the k directionThe projected point of (2) is,for a cluster close to the reference origin o,another cluster away from the reference origin o;is composed ofIs comprised of the projection points of the vertices of all planes perpendicular to k.
4. A spatial object cube component detection and estimation apparatus, comprising:
the acquisition module is used for acquiring a cubic component to be estimated;
the detection module is used for detecting the cube according to the relative spatial position between every two surfaces of the cube;
the estimation module is used for carrying out space attitude estimation and cube size estimation on the cube according to a robust estimation method;
the detection module comprises:
the judgment submodule is used for judging the position of the plane of the cube;
a determination submodule for determining a cube plane;
the judgment of the cube plane position is a plane pair { Pp,PqThe judgment of the relative spatial position comprises the position relationship of two cubic planes:
the opposite cubic plane discrimination condition: i) plane normal vector parallel, np||nq(ii) a ii) the edge vectors are correspondingly parallel, xp||xq,yp||yqOr xp||yq,yp||xq(ii) a iii) the planes are positioned opposite and of similar dimensions,
adjacent cube face discrimination conditions: i) plane normal vector is vertical: n isp⊥nq(ii) a ii) an edgeThe vectors corresponding to parallel or perpendicular, xp||xq,xp⊥yq,yp⊥xq,yp⊥yqOr yp||xq,yp⊥yq,xp⊥xq,xp⊥yq(ii) a iii) the planes are adjacently positioned and have similar dimensions, such that PpX ofpEdge and PqX ofqAdjacent, then need to satisfy
Wherein c is a reference vertex of plane P ═ Π (c, n, x, y), n is a unit normal vector of plane P, and x and y are edge vectors of plane P; subscripts p and q correspond to different planes; if (a, b)<TθThen consider a | | b, if |90 ° - < a, b-<TθConsidering a ^ b, wherein the angle (a, b) is the included angle between a and b; will plane PaTo plane PbProjection gave P'aThen P isaAnd PbHas a cross-over ratio ofS (-) is the planar area; dabIs a plane PaCenter point to plane PbVector D of center pointab=(ca+(xa+ya)/2)-(cb+(xb+yb)/2);
The determination of the cube plane comprises:
b) Set of traversal planesIn all plane pairs { Pp,PqIf { P }p,PqWhen the opposite cube face discrimination condition or the adjacent cube face discrimination condition is satisfied, theTaking the cubic plane as a reference cubic plane group and entering the next step; if not, then,there is no cubic plane present;
c) set of traversal planesAll planes P inmIf, ifPmAndall planes in (1) satisfy the opposite plane relation or the adjacent plane relation, andthe cube configuration is satisfied: one surface has only one opposite surface and four adjacent surfaces in different directions, P ismIs contained inNamely, it is
d) If it is notZhongzhuanping (Chinese character of 'Zhongzhuanping' for treating rheumatism)Number of faces greater thanNumber of opposing planes in, orNumber of median opposition planes equal toAnd the number of opposite planes inTotal number of midplane is greater thanTotal number of planes in (1) isReplacement of
e) If it is notIn which at least one set of opposing planes is present, orIf the total number of planes is not less than three, the number is considered to beIs an effective cubic plane set, pairTo carry outEstimation of the attitude of the cube space and estimation of the cube size, and fromIn the middle of removingTraversing the steps of a) to d) untilNo valid cube-plane groups are extracted.
5. The apparatus of claim 4, wherein the estimation module for spatial pose estimation of a cube comprises:
estimating the direction of the cubic edge vector according to the cubic plane, wherein the formula is as follows:
whereinIs a cube plane group, k belongs to { u, v, w } corresponding to the cube edge;andare respectively asA plane perpendicular to k and parallel to k; δ (·, ·) is an exemplary function, δ (a, b) equals 1 if a | | b, otherwise δ (a, b) equals 0; w (-) is a weight function, and for a plane normal vector n, W (n) is defined as the reciprocal of the root mean square of the total distance error from all plane points to the plane; for plane edge vector Is defined as a planar point edgeDegree of uniformity of directional distribution: point the plane toProjection, arranging the projection points in sequence and equally dividing the projection points into NbinSegment, then
Wherein s isiIs the proportion of the proxels falling within the ith segment.
6. The apparatus according to any of claims 4-5, wherein the estimation module is further configured to perform size estimation on the cube, comprising:
for input cube plane groupsAnd the estimated edge vectorEstimating cube in each directionLength of dimension (d);
estimating the length of each side of the cube according to the plane of the cube, wherein the calculation formula is as follows:
whereinEdges of the corresponding cube;is composed ofThe projection points from the top points of all the planes to k;andare two clusters distributed at two ends of the k directionThe projected point of (2) is,for a cluster close to the reference origin o,another cluster away from the reference origin o;is composed ofIs comprised of the projection points of the vertices of all planes perpendicular to k.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711445395.6A CN108038876B (en) | 2017-12-27 | 2017-12-27 | Method and device for detecting and estimating cubic component of space target |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711445395.6A CN108038876B (en) | 2017-12-27 | 2017-12-27 | Method and device for detecting and estimating cubic component of space target |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108038876A CN108038876A (en) | 2018-05-15 |
CN108038876B true CN108038876B (en) | 2022-02-11 |
Family
ID=62098036
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711445395.6A Active CN108038876B (en) | 2017-12-27 | 2017-12-27 | Method and device for detecting and estimating cubic component of space target |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108038876B (en) |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
FR2993371B1 (en) * | 2012-07-13 | 2014-08-15 | Thales Sa | OPTICAL ORIENTATION AND POSITION MEASUREMENT SYSTEM WITHOUT PICTURE SOURCE IMAGE FORMATION AND MASK |
CN106056128B (en) * | 2016-04-20 | 2019-03-22 | 北京航空航天大学 | Remote sensing image classification mask method based on composite diagram condition random field |
CN106780751A (en) * | 2017-01-19 | 2017-05-31 | 桂林电子科技大学 | Three-dimensional point cloud method for reconstructing based on improved shielding Poisson algorithm |
CN107024932A (en) * | 2017-03-01 | 2017-08-08 | 上海师范大学 | The control method of intelligent four-wheel drive serving trolley |
-
2017
- 2017-12-27 CN CN201711445395.6A patent/CN108038876B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN108038876A (en) | 2018-05-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Kendall et al. | Geometric loss functions for camera pose regression with deep learning | |
Yang et al. | Efficient 3d room shape recovery from a single panorama | |
Xu et al. | Reconstruction of scaffolds from a photogrammetric point cloud of construction sites using a novel 3D local feature descriptor | |
CN105993034B (en) | Contour completion for enhanced surface reconstruction | |
US10846844B1 (en) | Collaborative disparity decomposition | |
CN110135455A (en) | Image matching method, device and computer readable storage medium | |
Micusik et al. | Descriptor free visual indoor localization with line segments | |
Lee et al. | Automatic integration of facade textures into 3D building models with a projective geometry based line clustering | |
CN103077509A (en) | Method for synthesizing continuous and smooth panoramic video in real time by using discrete cubic panoramas | |
CN104318551B (en) | Gauss hybrid models point cloud registration method based on convex closure characteristic key | |
CN111524168A (en) | Point cloud data registration method, system and device and computer storage medium | |
Wiemann et al. | Automatic construction of polygonal maps from point cloud data | |
CN113689535B (en) | Building model generation method and device based on unmanned aerial vehicle image | |
CN111415420A (en) | Spatial information determination method and device and electronic equipment | |
CN114202632A (en) | Grid linear structure recovery method and device, electronic equipment and storage medium | |
CN116402976A (en) | Training method and device for three-dimensional target detection model | |
CN113628343B (en) | Three-dimensional grid combination processing method and device and storage medium | |
CN115018999A (en) | Multi-robot-cooperation dense point cloud map construction method and device | |
CN111402429B (en) | Scale reduction and three-dimensional reconstruction method, system, storage medium and equipment | |
CN114066999A (en) | Target positioning system and method based on three-dimensional modeling | |
CN108038876B (en) | Method and device for detecting and estimating cubic component of space target | |
CN109785421B (en) | Texture mapping method and system based on air-ground image combination | |
Kutulakos | Shape from the light field boundary | |
CN113111741B (en) | Assembly state identification method based on three-dimensional feature points | |
Dou et al. | Robust image matching based on SIFT and delaunay triangulation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |