CN111401385A - Similarity calculation method for image local topological structure feature descriptors - Google Patents

Similarity calculation method for image local topological structure feature descriptors Download PDF

Info

Publication number
CN111401385A
CN111401385A CN202010194853.9A CN202010194853A CN111401385A CN 111401385 A CN111401385 A CN 111401385A CN 202010194853 A CN202010194853 A CN 202010194853A CN 111401385 A CN111401385 A CN 111401385A
Authority
CN
China
Prior art keywords
points
corresponding relation
point
neighborhood
distance
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
CN202010194853.9A
Other languages
Chinese (zh)
Other versions
CN111401385B (en
Inventor
桑强
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.)
Chengdu Univeristy of Technology
Original Assignee
Chengdu Univeristy of Technology
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 Chengdu Univeristy of Technology filed Critical Chengdu Univeristy of Technology
Priority to CN202010194853.9A priority Critical patent/CN111401385B/en
Publication of CN111401385A publication Critical patent/CN111401385A/en
Application granted granted Critical
Publication of CN111401385B publication Critical patent/CN111401385B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching

Abstract

The invention provides a similarity calculation method of an image local topological structure feature descriptor, firstly, a local topological structure feature descriptor is provided, secondly, when the local similarity is compared, the corresponding relation of the local structures is found by two partial order structures based on the characteristics of the descriptor, and the difference between different local topological structures is quantified by the provided cost calculation method, so that the similarity of the topological structures can be effectively calculated. Compared with the existing method, the method accurately quantifies the difference of local structures.

Description

Similarity calculation method for image local topological structure feature descriptors
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to a similarity calculation method for an image local topological structure feature descriptor.
Background
The local features are used as a powerful tool and have been widely applied in computer vision systems and multiple application scenarios, wherein point features are the most important class of image features due to good feature invariance and computational real-time performance, and particularly in S L AM and stereo vision algorithms, whether local stability can be maintained during object matching, which affects registration accuracy and final 3D modelingThe key to correctness is that the image is represented in the form of a discrete point set by a feature point extraction algorithm. When calculating the local features of images, the currently mainstream local point feature calculation method is based on a shape context (shapecontext) and a similar histogram-based feature descriptor method, the shape context is mainly used for measuring the similarity of the contour shapes of two images, the statistical distribution characteristic of neighborhood points around a feature center point is utilized, and the shape features around the center point are described according to the angle and distance information of the distribution of the neighborhood points. Other similar methods include a contour point histogram (CPDH), which uses the center of the minimum circumscribed circle of the target object as the center point, forms a log-polar coordinate histogram according to the distribution of contour points, and calculates the similarity measure between two contour histograms using the EMD (Earth Mover's distance) distance when calculating the contour similarity. Most of recent algorithms, especially in 3D point cloud registration, shape context and similar methods are selected when calculating local features, however, these methods are mainly used to describe image contours, not local structures of targets, and shape context can describe local features of images to some extent. When describing a target local structure, due to the requirement of locality, sampling points are necessarily limited, usually several neighbor points around a central point, usually less than ten, are not enough to form an effective feature histogram, secondly, due to sparse histograms formed by too few feature points, local similarity cannot be effectively calculated, and when quantifying the similarity, the degree of similarity of different local features needs to be distinguished. But between calculating two histograms2In distance, the calculated cost values are the same as long as the contour points are distributed on different grids of the histogram. For example, three equal length sides with angles of 0 °, 45 °, 90 °, respectively, have a shape calculated from the context of the shape2The distances are the same, and in the general sense, it is considered that the 0 ° side is more similar to the 45 ° side, and the 90 ° side is more different, so that the method cannot accurately represent the degree of the profile difference.
Disclosure of Invention
In view of the above defects in the prior art, the similarity calculation method for the image local topological structure feature descriptors provided by the invention can effectively calculate the similarity of the topological structure, and accurately quantizes the difference of the local structure compared with the existing method.
In order to achieve the above purpose, the invention adopts the technical scheme that:
the scheme provides a similarity calculation method for image local topological structure feature descriptors, which comprises the following steps:
s1, representing the image contour as a set of points SPAnd according to said points, collecting points piObtaining N points nearest to the midpoint from the neighborhood points, and forming a neighborhood structure of the image by the N points;
s2, obtaining an angle value and a distance value between the two points of the neighborhood point and the central point;
s3, respectively obtaining a distance sequence and an angle sequence of the neighborhood points around the central point according to the angle value and the distance value, and forming a feature descriptor of the image local topological structure by the distance sequence and the angle sequence;
s4, when comparing similarity, obtaining an initial corresponding relation of a distance sequence according to the distance sequence based on the feature descriptor, and obtaining a corresponding relation of angle sequence verification according to the angle sequence;
s5, judging whether the initial corresponding relation is consistent with the verification relation, if so, adding the consistent corresponding relation into the final corresponding relation z, and entering the step S6, otherwise, entering the step S7;
s6, according to the final corresponding relation z, calculating by using a cost calculation method to obtain a matching probability P' when the initial corresponding relation is consistent with the verification corresponding relation;
s7, calculating by using a cost calculation method to obtain a matching probability P' when the initial corresponding relation is inconsistent with the verification corresponding relation;
s8, according to the matching probability P 'and the matching probability P' quantification structure difference, completing the similarity calculation of the image local topological structure feature descriptor.
Further, the feature descriptor in the step S3
Figure BDA0002417233750000031
The expression of (a) is as follows:
Figure BDA0002417233750000032
Figure BDA0002417233750000033
and is
Figure BDA0002417233750000034
Figure BDA0002417233750000035
And is
Figure BDA0002417233750000036
Wherein the content of the first and second substances,
Figure BDA0002417233750000037
a sequence of distances is represented which is,
Figure BDA0002417233750000038
is represented by piOne point of the set of distance sequences formed for the feature center, piRepresenting any one of the feature center points on the image contour, L (-) representing the length distance between two points,
Figure BDA0002417233750000039
a sequence of angles is represented which is,
Figure BDA00024172337500000310
is represented by piFor one point in the set of angular sequences formed for the feature center, θ (-) represents the angular distance between the two points.
Still further, step S4 is specifically:
based on the feature descriptors, sequentially corresponding the distances of the neighborhood points in the distance sequence from small to large one by one to obtain an initial corresponding relation of the distance sequence; and
and based on the feature descriptors, sequentially corresponding the angles of the neighborhood points in the angle sequence from small to large one by one to obtain the initial corresponding relation of the angle sequence.
Still further, the structure difference S quantified in the step S6mnThe expression of (a) is as follows:
Smn＝P”+P'
Figure BDA0002417233750000041
Figure BDA0002417233750000042
Figure BDA0002417233750000043
wherein, P 'represents the matching probability when the initial corresponding relation is inconsistent with the verification corresponding relation, P' represents the matching probability when the initial corresponding relation is consistent with the verification corresponding relation, z represents the final corresponding relation, k represents that k points in N neighborhood points are matched, N represents the number of the neighborhood points, t represents the matching probability when the initial corresponding relation is inconsistent with the verification corresponding relation, k represents the matching probability when the initial corresponding relation is consistent with the verification corresponding relation, k represents that k pointsiRepresenting a certain point in a set, djRepresenting a point in another set, x being denoted by tiForming a certain point in the neighborhood for the centre point, y being denoted by djForming a certain point in the neighborhood for the centre point, tmThe representation and x form the points in the neighborhood point pair,
Figure BDA0002417233750000044
is represented by tmForming points x, d in the neighborhood for the center pointnThe representation and x form the points in the neighborhood point pair,
Figure BDA0002417233750000045
is represented by tmA point y in the neighborhood is formed for the center point,
Figure BDA0002417233750000046
represents tiThe neighborhood of points x of the image is,
Figure BDA0002417233750000047
denotes djNeighborhood point y, liRepresents tiPoint to point distance,/jDenotes djPair distance of points,/mRepresenting the distance, l, of all pairs of points computed and composed of the neighborhood points xnRepresenting the distance, θ, of all pairs of points computed and made up of the neighborhood points yiRepresents tiPoint to angle of (theta)jDenotes djDiagonal angle of points, θmRepresenting the angle, θ, of all pairs of points computed and made up of the neighborhood points xnRepresenting the angles of all pairs of points computed and made up of the neighborhood points y,
Figure BDA0002417233750000048
and
Figure BDA0002417233750000049
respectively, the maximum angle and distance values between the pairs of points, and C (-) represents the cost difference between the two pairs of points.
The invention has the beneficial effects that:
(1) the invention provides a new local structure feature descriptor, which consists of two sequences of feature point neighborhoods: one sequence describes the distance arrangement relation of the neighborhood, the other describes the angle arrangement relation of the neighborhood, and the two spatial relations accurately describe the stability of the local structure. Based on the stability, the method can accurately describe the local topological structure of the points, has good characteristic invariance, and is not influenced by the sparsity of the point set, high in accuracy and good in real-time particularly compared with the traditional histogram statistical distribution method;
(2) the invention provides a similarity calculation method based on the feature descriptors. When comparing the local similarity, the corresponding relation between two adjacent domains is obtained by the local stable relation described by the two sequences in the descriptor, wherein an initial corresponding relation can be obtained according to the distance sequence, a verification corresponding relation is obtained according to the angle sequence, if the two relations are consistent, the one-to-one corresponding relation between the adjacent domains can be determined, otherwise, the characteristics are not matched, and finally, the local difference of the characteristics is quantized by the corresponding relation through the proposed point-to-point calculation method. Because the similarity calculation is based on the local corresponding relation, the quantification of the similarity is more accurate, and different local areas of the image can be accurately compared;
(3) the feature descriptor integrates the context information of the image feature points, quantifies the difference degree between the local features of the image through the prior corresponding relation between neighborhoods, and plays a key role in computer vision engineering as an image local feature quantification method.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
Detailed Description
The following description of the embodiments of the present invention is provided to facilitate the understanding of the present invention by those skilled in the art, but it should be understood that the present invention is not limited to the scope of the embodiments, and it will be apparent to those skilled in the art that various changes may be made without departing from the spirit and scope of the invention as defined and defined in the appended claims, and all matters produced by the invention using the inventive concept are protected.
Examples
As shown in fig. 1, the present invention provides a similarity calculation method for image local topology feature descriptors, which is implemented as follows:
s1, representing the image contour as a set of points SPAnd according to said points, collecting points piObtaining N points nearest to the midpoint from the neighborhood points, and forming a neighborhood structure of the image by the N points;
s2, obtaining an angle value and a distance value between the two points of the neighborhood point and the central point;
s3, respectively obtaining a distance sequence and an angle sequence of the neighborhood points around the central point according to the angle value and the distance value, and forming a feature descriptor of the image local topological structure by the distance sequence and the angle sequence;
s4, when comparing similarity, obtaining an initial corresponding relation of a distance sequence according to the distance sequence based on the feature descriptor, and obtaining a corresponding relation of angle sequence verification according to the angle sequence;
s5, judging whether the initial corresponding relation is consistent with the verification relation, if so, adding the consistent corresponding relation into the final corresponding relation z, and entering the step S6, otherwise, entering the step S7;
s6, according to the final corresponding relation z, calculating by using a cost calculation method to obtain a matching probability P' when the initial corresponding relation is consistent with the verification corresponding relation;
s7, calculating by using a cost calculation method to obtain a matching probability P' when the initial corresponding relation is inconsistent with the verification corresponding relation;
s8, according to the matching probability P 'and the matching probability P' quantification structure difference, completing the similarity calculation of the image local topological structure feature descriptor.
In this embodiment, first an image is composed of sample points on its contour, and then a contour can be represented as a set of points SP＝p1,p2…,pMWhere M is the potential of the set of points. Point p in the setiIs a neighborhood of
Figure BDA0002417233750000076
Is defined as the N points closest to the point, where o 1, 2. The local topological structure of the point is formed by the N points and can be a stable structure, the relation between the adjacent point and the central point is defined to be formed by an angle value and a distance value between the two points, the sum of the relation between the N point pairs uniquely determines the local topological structure of the central point, and the neighborhood points can form two ordered sets around the central point according to the size of the angle value and the distance value.
Figure BDA0002417233750000071
Wherein, satisfy
Figure BDA0002417233750000072
L (-) indicates the distance in length between two points, here L2Distance. In the same way, the method for preparing the composite material,
Figure BDA0002417233750000073
wherein, satisfy
Figure BDA0002417233750000074
θ (-) represents the angular distance between two points. The new local structure descriptor proposed by the present invention is composed of these two ordered sets. For point piThe descriptor
Figure BDA0002417233750000075
In this embodiment, when the contour point set is rigidly deformed, the descriptor is kept consistent because the distance and angular relationship between the point pairs is not changed. When the contour points are elastically deformed, the point set is subjected to anisotropic changes, i.e., the changes of the distance between the point pairs and the angle value are arbitrary. However, it must be of some significance for the elastic deformation of the target object encountered in engineering applications. For example, the deformation of internal organs during respiration, the deformation of outline of animal walking, and the change of outline of different handwritten Chinese characters. Although the shape changes of the different objects are elastic deformation, a certain rule is followed, namely the intrinsic meaning of the object is not changed after the object is elastically deformed, namely, although the shape of the heart changes during respiration change, the outline characteristics still indicate that the heart belongs to a heart organ or a cheetah during walking, and the changing shape characteristics can still be recognized by people at a glance as a cheetah. If the deformation is random, the target can not be identified completely, and the subsequent application value is lost, and no significance is realized. Therefore, based on the idea of differentiation, the image contour occurs as a wholeIs an elastic deformation, i.e. a change in anisotropy; locally, the variation between points is relatively smooth, approximating an isotropic variation, with the variation in angle and length between a point and its neighbors being consistent. That is, the arrangement sequence of the neighborhood points around the central point is stable, and after the whole contour is elastically deformed, the local points TpiThe two ordered sets in the topology descriptor of (1) are stable and invariant.
In this embodiment, when calculating the similarity between two local structures, if the distributions of the neighboring points are completely the same, we can consider that the similarity between the two local structures is the largest, and conversely, the similarity is the smallest. When the number of the neighborhood points is equal to one (N is 1), the local structure cost difference can be obtained only by calculating the angle difference and the distance difference between two point pairs, because only one neighborhood point is a corresponding point, and when N is equal to>When 2, the key point is how to determine the correspondence between the local structure neighborhood points. The invention solves the corresponding relation between local structures according to the topology descriptors described above, and because the point sequences of two sets in the descriptors keep stable when the whole point set is elastically deformed, the corresponding two descriptors for two unknown local structures
Figure BDA0002417233750000081
And
Figure BDA0002417233750000082
the determined correspondences are equally stable to the sequence. First, according to the descriptor
Figure BDA0002417233750000083
In (1)
Figure BDA0002417233750000084
And
Figure BDA0002417233750000085
in (1)
Figure BDA0002417233750000086
According to the distance valueAnd sorting the sizes of the two neighborhood point sets to obtain the corresponding relation between the two neighborhood point sets. Secondly, according to
Figure BDA0002417233750000087
In (1)
Figure BDA0002417233750000088
And
Figure BDA0002417233750000089
in (1)
Figure BDA00024172337500000810
The corresponding relation between two neighborhood point sets correspondingly obtained according to the sequence of the sorting of the angle values can be obtained, namely, the correspondence is one-to-one according to the sequence of the distance of the neighborhood points and the angle values from small to large, namely
Figure BDA00024172337500000811
Corresponding to a point
Figure BDA00024172337500000812
Or is provided with
Figure BDA00024172337500000813
Corresponding to a point
Figure BDA00024172337500000814
And the two obtained correspondences are respectively recorded as: z is a radical oflAnd zθIf two points are at zlAnd zθAre consistent, then the two points are marked as corresponding points, if all the points in the two local structures satisfy the above correspondence, then the two local structures can be considered completely similar, otherwise, they are not similar or partially similar. Let k points in the partial structure correspond to each other, and the correspondence is taken as z. And meanwhile, N-k points are not matched, and for the matched point pairs, the quantized local structure similarity probability is given as follows:
Figure BDA00024172337500000815
for a point pair that is an unmatched point pair, the matching probability is:
Figure BDA0002417233750000091
here, C (·) is the cost difference between two pairs of points:
Figure BDA0002417233750000092
in the above formula, the first and second carbon atoms are,
Figure BDA0002417233750000093
and
Figure BDA0002417233750000094
respectively representing the maximum distance and angle values between the point pairs, wherein the final similarity calculation probability is as follows: smn＝P”+P'。
In summary, the present invention proposes a new local structure descriptor, which is composed of two sequences of feature point neighborhoods: one sequence describes the distance arrangement relation of the neighborhoods, the other describes the angle arrangement relation of the neighborhoods, the two spatial relations accurately describe the stability of the local structure, and then the corresponding relation between the two neighborhoods can be obtained based on the local stability relation described by the two sequences, wherein an initial corresponding relation can be obtained according to the distance sequence, a verification corresponding relation can be obtained according to the angle sequence, and if the two relations are consistent, the corresponding relation between the neighborhoods is obtained. Otherwise, the features are not matched, and finally, the quantized structural difference is calculated according to the obtained corresponding relation.

Claims (4)

1. A similarity calculation method for image local topological structure feature descriptors is characterized by comprising the following steps:
s1, contour the imageThe lines are represented as a set of points SPAnd according to said points, collecting points piObtaining N points nearest to the midpoint from the neighborhood points, and forming a neighborhood structure of the image by the N points;
s2, obtaining an angle value and a distance value between the two points of the neighborhood point and the central point;
s3, respectively obtaining a distance sequence and an angle sequence of the neighborhood points around the central point according to the angle value and the distance value, and forming a feature descriptor of the image local topological structure by the distance sequence and the angle sequence;
s4, when comparing the similarity, based on the feature descriptor, obtaining the initial corresponding relation of the distance sequence according to the distance sequence, and obtaining the verification corresponding relation of the angle sequence according to the angle sequence;
s5, judging whether the initial corresponding relation is consistent with the verification relation, if so, adding the consistent corresponding relation into the final corresponding relation z, and entering the step S6, otherwise, entering the step S7;
s6, according to the final corresponding relation z, calculating by using a cost calculation method to obtain a matching probability P' when the initial corresponding relation is consistent with the verification corresponding relation;
s7, calculating by using a cost calculation method to obtain a matching probability P' when the initial corresponding relation is inconsistent with the verification corresponding relation;
s8, according to the matching probability P 'and the matching probability P' quantification structure difference, completing the similarity calculation of the image local topological structure feature descriptor.
2. The method for calculating similarity of image local topological structure feature descriptors according to claim 1, wherein said feature descriptors in step S3
Figure FDA0002417233740000011
The expression of (a) is as follows:
Figure FDA0002417233740000012
Figure FDA0002417233740000013
and is
Figure FDA0002417233740000014
Figure FDA0002417233740000015
And is
Figure FDA0002417233740000016
Wherein the content of the first and second substances,
Figure FDA0002417233740000021
a sequence of distances is represented which is,
Figure FDA0002417233740000022
is represented by piOne point of the set of distance sequences formed for the feature center, piRepresenting any one of the feature center points on the image contour, L (-) representing the length distance between two points,
Figure FDA0002417233740000023
a sequence of angles is represented which is,
Figure FDA0002417233740000024
is represented by piFor one point in the set of angular sequences formed for the feature center, θ (-) represents the angular distance between the two points.
3. The method for calculating the similarity of the image local topological structure feature descriptors according to claim 1, wherein the step S4 specifically includes:
based on the feature descriptors, sequentially corresponding the distances of the neighborhood points in the distance sequence from small to large one by one to obtain an initial corresponding relation of the distance sequence; and
and based on the feature descriptors, sequentially corresponding the angles of the neighborhood points in the angle sequence from small to large one by one to obtain the initial corresponding relation of the angle sequence.
4. The method for calculating similarity of image local topological structure feature descriptors according to claim 1, wherein said structure difference S quantified in step S8 is SmnThe expression of (a) is as follows:
Smn＝P”+P'
Figure FDA0002417233740000025
Figure FDA0002417233740000026
Figure FDA0002417233740000027
wherein, P 'represents the matching probability when the initial corresponding relation is inconsistent with the verification corresponding relation, P' represents the matching probability when the initial corresponding relation is consistent with the verification corresponding relation, z represents the final corresponding relation, k represents that k points in N neighborhood points are matched, N represents the number of the neighborhood points, t represents the matching probability when the initial corresponding relation is inconsistent with the verification corresponding relation, k represents the matching probability when the initial corresponding relation is consistent with the verification corresponding relation, k represents that k pointsiRepresenting a certain point in a set, djRepresenting a point in another set, x being denoted by tiForming a certain point in the neighborhood for the centre point, y being denoted by djForming a certain point in the neighborhood for the centre point, tmThe representation and x form the points in the neighborhood point pair,
Figure FDA0002417233740000031
is represented by tmForming points x, d in the neighborhood for the center pointnThe representation and x form the points in the neighborhood point pair,
Figure FDA0002417233740000032
is represented by tmA point y in the neighborhood is formed for the center point,
Figure FDA0002417233740000033
represents tiThe neighborhood of points x of the image is,
Figure FDA0002417233740000034
denotes djNeighborhood point y, liRepresents tiPoint to point distance,/jDenotes djPair distance of points,/mRepresenting the distance, l, of all pairs of points computed and composed of the neighborhood points xnRepresenting the distance, θ, of all pairs of points computed and made up of the neighborhood points yiRepresents tiPoint to angle of (theta)jDenotes djDiagonal angle of points, θmRepresenting the angle, θ, of all pairs of points computed and made up of the neighborhood points xnRepresenting the angles of all pairs of points computed and made up of the neighborhood points y,
Figure FDA0002417233740000035
and
Figure FDA0002417233740000036
respectively, the maximum angle and distance values between the pairs of points, and C (-) represents the cost difference between the two pairs of points.
CN202010194853.9A 2020-03-19 2020-03-19 Similarity calculation method for image local topological structure feature descriptors Expired - Fee Related CN111401385B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010194853.9A CN111401385B (en) 2020-03-19 2020-03-19 Similarity calculation method for image local topological structure feature descriptors

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010194853.9A CN111401385B (en) 2020-03-19 2020-03-19 Similarity calculation method for image local topological structure feature descriptors

Publications (2)

Publication Number Publication Date
CN111401385A true CN111401385A (en) 2020-07-10
CN111401385B CN111401385B (en) 2022-06-17

Family

ID=71436972

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010194853.9A Expired - Fee Related CN111401385B (en) 2020-03-19 2020-03-19 Similarity calculation method for image local topological structure feature descriptors

Country Status (1)

Country Link
CN (1) CN111401385B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117235545A (en) * 2023-11-14 2023-12-15 亿海蓝（北京）数据技术股份公司 Method, device and readable storage medium for judging spatial object distribution similarity

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030044048A1 (en) * 2001-06-18 2003-03-06 Zhengyou Zhang Incremental motion estimation through local bundle adjustment
US8520949B1 (en) * 2008-06-20 2013-08-27 Google Inc. Self-similar descriptor filtering
US20140355832A1 (en) * 2011-12-02 2014-12-04 Morpho Method and Device for Following an Object in a Sequence of at Least Two Images
CN104835116A (en) * 2015-04-10 2015-08-12 山东师范大学 Contour-based two-dimensional fragment splicing method
CN105405116A (en) * 2014-09-04 2016-03-16 南京理工大学 Stereo matching method based on graph cut
CN106530341A (en) * 2016-11-01 2017-03-22 成都理工大学 Point registration algorithm capable of keeping local topology invariance
CN106530338A (en) * 2016-10-31 2017-03-22 武汉纺织大学 Method and system for matching MR image feature points before and after nonlinear deformation of biological tissue
CN107808362A (en) * 2017-11-15 2018-03-16 北京工业大学 A kind of image split-joint method combined based on unmanned plane POS information with image SURF features
CN108416732A (en) * 2018-02-02 2018-08-17 重庆邮电大学 A kind of Panorama Mosaic method based on image registration and multi-resolution Fusion
CN109472816A (en) * 2018-09-17 2019-03-15 西北大学 A kind of point cloud registration method
CN109948624A (en) * 2019-02-18 2019-06-28 北京旷视科技有限公司 Method, apparatus, electronic equipment and the computer storage medium of feature extraction

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030044048A1 (en) * 2001-06-18 2003-03-06 Zhengyou Zhang Incremental motion estimation through local bundle adjustment
US8520949B1 (en) * 2008-06-20 2013-08-27 Google Inc. Self-similar descriptor filtering
US20140355832A1 (en) * 2011-12-02 2014-12-04 Morpho Method and Device for Following an Object in a Sequence of at Least Two Images
CN105405116A (en) * 2014-09-04 2016-03-16 南京理工大学 Stereo matching method based on graph cut
CN104835116A (en) * 2015-04-10 2015-08-12 山东师范大学 Contour-based two-dimensional fragment splicing method
CN106530338A (en) * 2016-10-31 2017-03-22 武汉纺织大学 Method and system for matching MR image feature points before and after nonlinear deformation of biological tissue
CN106530341A (en) * 2016-11-01 2017-03-22 成都理工大学 Point registration algorithm capable of keeping local topology invariance
CN107808362A (en) * 2017-11-15 2018-03-16 北京工业大学 A kind of image split-joint method combined based on unmanned plane POS information with image SURF features
CN108416732A (en) * 2018-02-02 2018-08-17 重庆邮电大学 A kind of Panorama Mosaic method based on image registration and multi-resolution Fusion
CN109472816A (en) * 2018-09-17 2019-03-15 西北大学 A kind of point cloud registration method
CN109948624A (en) * 2019-02-18 2019-06-28 北京旷视科技有限公司 Method, apparatus, electronic equipment and the computer storage medium of feature extraction

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
ERION HASANBELLIU ET AL: "Information Theoretic Shape Matching", 《 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE》 *
J. RABIN ET AL: "A Statistical Approach to the Matching of Local Features", 《SIAM JOURNAL ON IMAGING SCIENCES》 *
QIANG SANG ET AL: "Robust non-rigid point registration based on feature-dependant finite mixture model", 《PATTERN RECOGNITION LETTERS》 *
YONG-XING HE ET AL: "Application of neighborhood feature in point clouds registration", 《 PROCEEDINGS OF 2011 INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND NETWORK TECHNOLOGY》 *
杨燕: "复杂场景图像特征分析与匹配方法研究", 《中国博士学位论文全文数据库 信息科技辑》 *
翟凤文: "图像特征点匹配关键技术研究", 《中国博士学位论文全文数据库 信息科技辑》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117235545A (en) * 2023-11-14 2023-12-15 亿海蓝（北京）数据技术股份公司 Method, device and readable storage medium for judging spatial object distribution similarity
CN117235545B (en) * 2023-11-14 2024-02-02 亿海蓝（北京）数据技术股份公司 Method, device and readable storage medium for judging spatial object distribution similarity

Also Published As

Publication number Publication date
CN111401385B (en) 2022-06-17

Similar Documents

Publication Publication Date Title
CN110097093B (en) Method for accurately matching heterogeneous images
CN101980250B (en) Method for identifying target based on dimension reduction local feature descriptor and hidden conditional random field
CN109215129B (en) Local feature description method based on three-dimensional point cloud
CN106682700B (en) Block rapid matching method based on key point description operator
CN110807473B (en) Target detection method, device and computer storage medium
CN110334762B (en) Feature matching method based on quad tree combined with ORB and SIFT
CN111795704A (en) Method and device for constructing visual point cloud map
CN109697692B (en) Feature matching method based on local structure similarity
CN110807781B (en) Point cloud simplifying method for retaining details and boundary characteristics
CN104834931A (en) Improved SIFT algorithm based on wavelet transformation
CN112200915B (en) Front-back deformation detection method based on texture image of target three-dimensional model
CN107180436A (en) A kind of improved KAZE image matching algorithms
KR101182683B1 (en) A Visual Shape Descriptor Generating Method Using Sectors and Shape Context of Contour Lines and the Recording Medium thereof
CN108182705A (en) A kind of three-dimensional coordinate localization method based on machine vision
CN111199558A (en) Image matching method based on deep learning
CN114782715B (en) Vein recognition method based on statistical information
CN114332172A (en) Improved laser point cloud registration method based on covariance matrix
CN111401385B (en) Similarity calculation method for image local topological structure feature descriptors
CN112767429B (en) Ground-snow surface point cloud rapid segmentation method
CN114494380A (en) Binary shape context feature descriptor construction method and point cloud registration method
CN114358166A (en) Multi-target positioning method based on self-adaptive k-means clustering
Srivastava et al. Drought stress classification using 3D plant models
CN104392209A (en) Evaluation model for image complexity of target and background
CN109887012B (en) Point cloud registration method combined with self-adaptive search point set
CN117132630A (en) Point cloud registration method based on second-order spatial compatibility measurement

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
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20220617

CF01 Termination of patent right due to non-payment of annual fee