CN106529607A - Method and device for acquiring homonymy points of images - Google Patents
Method and device for acquiring homonymy points of images Download PDFInfo
- Publication number
- CN106529607A CN106529607A CN201611125944.7A CN201611125944A CN106529607A CN 106529607 A CN106529607 A CN 106529607A CN 201611125944 A CN201611125944 A CN 201611125944A CN 106529607 A CN106529607 A CN 106529607A
- Authority
- CN
- China
- Prior art keywords
- image
- characteristic point
- matching characteristic
- pair
- point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/757—Matching configurations of points or features
Abstract
The invention provides a method and a device for acquiring the homonymy points of images and belongs to the field of image processing. The method comprises the steps of acquiring a plurality of pairs of matched feature points which are mutually matched in a first image and a second image; calculating a plurality of corresponding first distances between each matched feature point in the first image and all other matched feature points in the first image, and calculating a plurality of corresponding second distances between each matched feature point in the second image and all other matched feature points in the second image; acquiring the distance difference between each first distance and a target second distance among the corresponding distances of each pair of matched feature points; calculating the average value of a plurality of distance differences corresponding to the plurality of pairs of matched feature points; deleting mutually matched feature points with the corresponding average value of the distance differences thereof meeting a preset distance deletion condition so as to acquire a plurality of pairs of matched feature points, and adopting the plurality of pairs of matched feature points as a plurality of pairs of matched homonymy points in the two to-be-matched images. Based on the method and the device, mutually matched feature points are further subjected to deletion treatment, so that finally acquired matched homonymy points are more accurate.
Description
Technical field
The present invention relates to image processing field, in particular to the same place acquisition methods and device of a kind of image.
Background technology
In multiple fields, the same place being mutually matched in often needing to obtain two width images such as unmanned plane is photogrammetric.It is existing
In technology, after the characteristic point being mutually matched directly is obtained using Feature Points Matching algorithm, the feature being mutually matched that will be obtained
O'clock as two width images same place.
But, in the characteristic point being mutually matched that directly matching is obtained, there may be the inaccurate error hiding of matching result
Point.
The content of the invention
In view of this, the same place acquisition methods and device of a kind of image are embodiments provided, by obtain
Further deleted with the Mismatching point in characteristic point, make the same place being mutually matched of acquisition more accurate.
To achieve these goals, the technical solution used in the present invention is as follows：
A kind of same place acquisition methods of image, methods described include：Obtain mutual in the first image and the second image
The multipair matching characteristic point matched somebody with somebody, described first image and the second image are two width images to be matched；Calculate described first image
In each matching characteristic point to corresponding multiple first distances of other all matching characteristic points, calculate each in second image
Matching characteristic point is to the corresponding multiple second distances of other all matching characteristic points；Obtain the corresponding distance of each pair matching characteristic point
In, each first distance is poor with the distance between target second distance, obtains multiple range differences of correspondence each pair matching characteristic point,
Wherein, two matching characteristic points of two matching characteristic points, first distance corresponding with formation of the target second distance are formed
Difference Corresponding matching；The meansigma methodss of the corresponding multiple range differences of each pair matching characteristic point are calculated, each pair matching is obtained
The range difference meansigma methodss of Feature point correspondence；The range difference meansigma methodss for meeting predeterminable range and deleting condition corresponding are mutually matched
Matching characteristic point deletion, the multipair matching same place of the multipair matching characteristic point of acquisition as two width images to be matched.
A kind of same place acquisition methods of image, methods described include：Obtain mutual in the first image and the second image
The multipair matching characteristic point matched somebody with somebody, described first image and the second image are two width images to be matched；With in described first image
Each characteristic point as starting point, other characteristic points form multiple primary vectors of each characteristic point of correspondence as terminal, with institute
Each characteristic point in the second image is stated as starting point, other characteristic points form multiple the of correspondence each characteristic point as terminal
Two is vectorial；Obtain in the corresponding vector of each pair matching characteristic point, the target in each primary vector and the plurality of secondary vector
The angle value of the angle between secondary vector, it is a pair with the terminal of corresponding primary vector that the target secondary vector is terminal
The secondary vector of matching characteristic point, obtains multiple angle values of correspondence each pair matching characteristic point；Calculate each pair matching characteristic
The meansigma methodss of the corresponding multiple angle values of point, obtain the corresponding angular average of each pair matching characteristic point；It is default by meeting
Angle deletes corresponding at least one pair of matching characteristic point deletion of angular average of condition, the multipair matching characteristic point conduct of acquisition
The multipair matching same place of two width images to be matched.
A kind of same place acquisition device of image, described device include：Match point acquisition module, for obtaining the first image
With the multipair matching characteristic point being mutually matched in the second image, described first image and the second image are two width figures to be matched
Picture；Distance calculation module, it is corresponding to other all matching characteristic points for calculating each matching characteristic point in described first image
Multiple first distances, in calculating second image, each matching characteristic point is corresponding multiple to other all matching characteristic points
Second distance；Range difference computing module, for obtaining in the corresponding distance of each pair matching characteristic point, each first distance and target
The distance between second distance is poor, obtains multiple range differences of correspondence each pair matching characteristic point, wherein, form the target second
Two matching characteristic point difference Corresponding matchings of two matching characteristic points, first distance corresponding with formation of distance；Range difference is put down
Mean value computation module, for calculating the meansigma methodss of the corresponding multiple range differences of each pair matching characteristic point, obtains each pair
The corresponding range difference meansigma methodss of matching characteristic point；First removing module, for the range difference by predeterminable range deletion condition is met
The corresponding matching characteristic point deletion being mutually matched of meansigma methodss, the multipair matching characteristic point of acquisition is used as two width images to be matched
Multipair matching same place.
A kind of same place acquisition device of image, described device include：Match point acquisition module, for obtaining the first image
With the multipair matching characteristic point being mutually matched in the second image, described first image and the second image are two width figures to be matched
Picture；Vectorial acquisition module, for using each characteristic point in described first image as starting point, other characteristic points are used as terminal shape
Into the multiple primary vectors for corresponding to each characteristic point, using each characteristic point in second image as starting point, other features
Point forms multiple secondary vectors of each characteristic point of correspondence as terminal；Angle calculation module, for obtaining each pair matching characteristic
In the corresponding vector of point, the angle of the angle between target secondary vector in each primary vector and the plurality of secondary vector
Value, the target secondary vector are terminal and the secondary vector that the terminal of corresponding primary vector is a pair of matching characteristic points, are obtained
Multiple angle values of each pair matching characteristic point must be corresponded to；Angular average computing module, for calculating each pair matching characteristic
The meansigma methodss of the corresponding multiple angle values of point, obtain the corresponding angular average of each pair matching characteristic point；Second deletes mould
Block, for corresponding at least one pair of the matching characteristic point deletion of angular average by default angle deletion condition is met, acquisition
Multipair matching same place of the multipair matching characteristic point as two width images to be matched.
The same place acquisition methods of image provided in an embodiment of the present invention and device, are obtaining two width images to be matched
After matching characteristic point, the matching characteristic point to wherein meeting deletion condition is deleted, with delete error hiding probability it is higher
With point, using deletion after the matching characteristic point being mutually matched as the same place being mutually matched of two width images to be matched, obtain
The same place accuracy rate being mutually matched for obtaining is higher.
To enable the above objects, features and advantages of the present invention to become apparent, preferred embodiment cited below particularly, and coordinate
Appended accompanying drawing, is described in detail below.
Description of the drawings
To make purpose, technical scheme and the advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
The a part of embodiment of the present invention, rather than the embodiment of whole.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Fig. 1 shows the block diagram of the computer that present pre-ferred embodiments are provided；
Fig. 2 shows a kind of flow chart of the same place acquisition methods of the image that first embodiment of the invention is provided；
The flow chart that Fig. 3 shows the same place acquisition methods of the image that second embodiment of the invention is provided；
Fig. 4 shows the structured flowchart of the same place acquisition device of the image that third embodiment of the invention is provided；
Fig. 5 shows the structured flowchart of the same place acquisition device of the image that fourth embodiment of the invention is provided.
Specific embodiment
Below in conjunction with accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Ground description, it is clear that described embodiment is only a part of embodiment of the invention, rather than the embodiment of whole.Generally exist
The component of the embodiment of the present invention described and illustrated in accompanying drawing can be arranged and be designed with a variety of configurations herein.Cause
This, the detailed description of the embodiments of the invention to providing in the accompanying drawings is not intended to limit claimed invention below
Scope, but it is merely representative of the selected embodiment of the present invention.Based on embodiments of the invention, those skilled in the art are not doing
The every other embodiment obtained on the premise of going out creative work, belongs to the scope of protection of the invention.
It should be noted that：Similar label and letter represent similar terms in following accompanying drawing, therefore, once a certain Xiang Yi
It is defined in individual accompanying drawing, then in subsequent accompanying drawing which further need not be defined and is explained.Meanwhile, the present invention's
In description, term " first ", " second " etc. are only used for distinguishing description, and it is not intended that indicating or implying relative importance.
As shown in figure 1, being the block diagram of the computer 100 that present pre-ferred embodiments are provided.The computer 100
Same place acquisition device, memorizer 101, storage control 102, processor 103, Peripheral Interface 104, input including image is defeated
Go out unit 105, display unit 106 and other.
It is the memorizer 101, storage control 102, processor 103, Peripheral Interface 104, input-output unit 105, aobvious
Show that 106 each element of unit is directly or indirectly electrically connected with each other, to realize the transmission or interaction of coordinate data.For example,
These elements can pass through one or more communication bus each other or holding wire is realized being electrically connected with.The same place of described image
Acquisition device includes at least one software that can be stored in the form of software or firmware (firmware) in the memorizer 101
Functional module.The processor 103 be used to perform memorizer 101 in the executable module that stores, such as described image it is of the same name
Software function module or computer program that point acquisition device includes.
Wherein, memorizer 101 may be, but not limited to, random access memory 101 (Random Access Memory,
RAM), read only memory 101 (Read Only Memory, ROM), 101 (Programmable of programmable read only memory
Read-Only Memory, PROM), 101 (Erasable Programmable Read-Only of erasable read-only memory
Memory, EPROM), 101 (Electric Erasable Programmable Read-Only of electricallyerasable ROM (EEROM)
Memory, EEPROM) etc..Wherein, memorizer 101 be used for storage program, the processor 103 after execute instruction is received,
Described program is performed, the side performed by the server/computer of the stream process definition that embodiment of the present invention any embodiment is disclosed
Method is can apply in processor 103, or is realized by processor 103.
A kind of possibly IC chip of processor 103, the disposal ability with signal.Above-mentioned processor 103 can
To be general processor 103, including central processing unit 103 (Central Processing Unit, abbreviation CPU), network processes
Device 103 (Network Processor, abbreviation NP) etc.；Can also be digital signal processor 103 (DSP), special IC
(ASIC), ready-made programmable gate array (FPGA) or other PLDs, discrete gate or transistor logic,
Discrete hardware components.Can realize or perform disclosed each method in the embodiment of the present invention, step and logic diagram.It is general
Processor 103 can be microprocessor 103 or the processor 103 can also be any conventional processor 103 etc..
Various input/output devices are coupled to processor 103 and memorizer 101 by the Peripheral Interface 104.At some
In embodiment, Peripheral Interface 104, processor 103 and storage control 102 can be realized in one single chip.Other one
In a little examples, they can be realized by independent chip respectively.
Input-output unit 105 is supplied to user input data to realize interacting for user and computer, is such as used for being input into many
Two field picture, so as to process to the multiple image, and exports result.The input-output unit can be, but not limit
In mouse and keyboard etc..
Display unit 106 provide between the computer and user an interactive interface (such as user interface) or
Refer to user for display image data.In the present embodiment, the display unit can be that liquid crystal display or touch-control are aobvious
Show device.If touch control display, which can be the capacitance type touch control screen or resistance type touch control screen for supporting single-point and multi-point touch operation
Deng.Support that single-point and multi-point touch operation refer to that touch control display can sense one or more positions on the touch control display
The touch control operation of place's simultaneously generation is put, and transfers to processor to be calculated and processed the touch control operation for sensing.
First embodiment
A kind of same place acquisition methods of image are present embodiments provided, it is same for what is be mutually matched in two width images of acquisition
Name point.Fig. 2 is referred to, the method includes：
Step S110：The multipair matching characteristic point being mutually matched in obtaining the first image and the second image, first figure
As being two width images to be matched with the second image.
First, Feature Points Matching is carried out to two width images to be matched, to obtain the multipair matching characteristic point being mutually matched.
In the present embodiment, the multipair matching characteristic being mutually matched in the first image and the second image can be obtained by sift algorithms
Point.
Specifically, first, to the first image and the second picture construction DOG metric space, can be entered by equation below
Row builds：
L (x, y, σ)=G (x, y, σ) * I (x, y)
D (x, y, σ)=(G (x, y, k σ)-G (x, y, σ)) * I (x, y)=L (x, y, k σ)-L (x, y, σ)
G (x, y, σ) is the Gaussian function of changeable scale.Wherein, σ is scale factor, and its value size represents that image is put down
The height of slippage degree, can be determined according to actual needs by user.(x, y) represents pixel coordinate in the picture.
To each image, extremum extracting is carried out in the metric space set up, in primarily determining that image, characteristic point institute is in place
Put and yardstick, the characteristic point of acquisition is discrete space extreme point.
Again the characteristic point of acquisition is fitted three-dimensional quadratic function accurately to determine position and the yardstick of characteristic point, it is low right to remove
Than the characteristic point and unstable skirt response point of degree.
Characteristic point direction can determine according to the principal direction of the feature vertex neighborhood gradient for obtaining, further according to the characteristic point for obtaining
Position, yardstick and principal direction, set up feature point description symbol.The descriptor of 128 dimensional feature vectors can be obtained.Pass through two again
The feature point description symbol of the characteristic point in width image carries out the calculating of Euclidean distance, obtains mutual in two width images to be matched
The multipair matching characteristic point matched somebody with somebody.
Certainly, in the present embodiment, the mode for obtaining the first image and the matching characteristic point in the second image is not made
To limit, it is also possible to obtained by other algorithms, such as SURF algorithm etc..
Further, before step S110, can also include carrying out noise filtering to the first image and the second image
And feature strengthens, wallis filtering algorithms can be passed through or bilateral filtering algorithm is realized.It is of course also possible to by other calculations
Method is realized, is not intended as in the present embodiment limiting.
Step S120：In calculating described first image, each matching characteristic point is corresponding more to other all matching characteristic points
Individual first distance, in calculating second image, each matching characteristic point is to other all matching characteristic points corresponding multiple second
Distance.
In step S110, multipair matching characteristic point { (P is obtainedi,Pi') | i=1,2 ..., n }, wherein, PiBelong to the first figure
Picture, Pi' belonging to the second image, n is the logarithm of the matching characteristic point of the first image and the second image.
Thus it is possible to understand, the first image includes multiple matching characteristic point Pi(i=1,2 ... n), in the second image
Including multiple matching characteristic point Pi' (i=1,2 ... n), multiple in the multiple matching characteristic points in the first image and the second image
Matching characteristic point distinguishes Corresponding matching.
Each the matching characteristic point in multiple matching characteristic points of the first image is calculated to other all matching characteristic points pair
Multiple first distances answered, first is as follows apart from computing formula：
As i=1, j is respectively 1,2,3 ... n obtains n the first distanceAs i=2, j is respectively 1,2 ... n,
Obtain n the first distanceBy that analogy, each the matching characteristic point in the first image of correspondence, obtains n the first distance
Likewise, each matching characteristic point is to other all matching characteristics in multiple matching characteristic points of the second image of calculating
The corresponding multiple second distances of point, second distance computing formula are as follows：
As i=1, j is respectively 1,2,3 ... n obtains n second distanceAs i=2, j is respectively 1,2,3,
4 ... n, obtain n second distanceBy that analogy, each the matching characteristic point in the second image of correspondence, obtains n second
Distance
Then, the corresponding distance of each pair matching characteristic point is n the first distanceAnd n second distance
Step S130：Obtain in the corresponding distance of each pair matching characteristic point, each first distance and target second distance it
Between range difference, obtain multiple range differences of correspondence each pair matching characteristic point, wherein, form two of the target second distance
Two matching characteristic point difference Corresponding matchings of matching characteristic point the first distance corresponding with formation.
For each pair characteristic point, n range difference of acquisition can be calculatedFor example, as i=1,
For i-th pair matching characteristic point, computed range is poorJ is respectively 1,2,3 ... n, n distance of acquisition
Difference dj.Likewise, as i=2, for i-th pair matching characteristic point, computed range is poorJ is respectively
1,2,3 ... n, obtains n range difference dj.For n is to matching characteristic point, n range difference d can be obtainedj(j=1,2,3 ... n).
In n range difference of each characteristic point, with the first distanceMiddle i and j equal second distanceFor
The target second distance of first distance, for example,ForTarget second distance,ForTarget second
Distance,ForTarget second distance etc..
Step S140：The meansigma methodss of the corresponding multiple range differences of each pair matching characteristic point are calculated, each pair is obtained
The corresponding range difference meansigma methodss of matching characteristic point.
For i-th pair matching characteristic point, it is possible to obtain the corresponding multiple n range differences of i-th pair matching characteristic point it is flat
Average, computing formula can be：
In can be to obtain corresponding range difference meansigma methodss AverDis of each pair matching characteristic pointi, for { (Pi, Pi') | i=
1,2, n } n obtains n range difference meansigma methodss AverDis to matching characteristic point, altogetheri(i=1,2,3 ... n).
Step S150：The corresponding matching characteristic being mutually matched of range difference meansigma methodss of predeterminable range deletion condition will be met
Point deletion, the multipair matching same place of the multipair matching characteristic point of acquisition as two width images to be matched.
In the present embodiment, the predeterminable range is deleted in all range difference meansigma methodss that condition can be acquisition, pre- definite proportion
Example RdMaximum range difference meansigma methodss.Predetermined ratio RdCan be determined according to practical situation by user, it is preferred that in this reality
Apply in example, can be 0.15.
Specifically, calculate in n range difference meansigma methodss, the logarithm for needing the matching characteristic point deleted is n*Rd.Again by n
In range difference meansigma methodss, maximum n*RdThe corresponding matching point deletion of individual range difference meansigma methodss.Corresponding of range difference meansigma methodss
It is acquisition range difference meansigma methodss AverDis with pointiA pair of matching characteristic points, for example, the meansigma methodss of i-th pair matching characteristic pointFor n*RdOne in individual maximum range difference meansigma methodss, then delete the i-th pair and match
Characteristic point.
Can be by being ranked up according to size order to n range difference meansigma methodss, deletion comes larger n*Rd
Individual maximum range difference meansigma methodss.Multipair matching the same place { (P of two width images to be matched can finally be obtainedi, Pi') | i=
1,2 ..., l }, PiFor the matching same place in the first image, Pi' it is the same place being mutually matched with Pi in the second image, can
With what is understood, l is less than n.
Second embodiment
A kind of same place acquisition methods of image are present embodiments provided, Fig. 3 is referred to, the method includes：
Step S210：The multipair matching characteristic point being mutually matched in obtaining the first image and the second image, first figure
As being two width images to be matched with the second image.
This step refer to step S110, will not be described here.
Step S220：Using each characteristic point in described first image as starting point, other characteristic points are formed as terminal
Multiple primary vectors of corresponding each characteristic point, using each characteristic point in second image as starting point, other characteristic points
Multiple secondary vectors of each characteristic point of correspondence are formed as terminal.
In step S110, multipair matching characteristic point { (P is obtainedi,Pi') | i=1,2 ..., n } wherein, PiBelong to the first figure
Picture, Pi' belonging to the second image, n is the logarithm of the matching characteristic point of the first image and the second image.
Thus it is possible to understand, the first image includes multiple matching characteristic point Pi(i=1,2 ... n), in the second image
Including multiple matching characteristic point Pi' (i=1,2 ... n), multiple in the multiple matching characteristic points in the first image and the second image
Matching characteristic point distinguishes Corresponding matching.
Each characteristic point in the first image is obtained as starting point, other characteristic points are accordingly special as the correspondence that terminal is formed
Levy multiple primary vectors a little：
I.e. with i as starting point, other characteristic points j=1,2,3 ... n are terminal, form n primary vectorFor example, work as i=
When 1, j is respectively 1,2,3 ... n obtains n primary vectorAs i=2, j is respectively 1,2 ... n, obtain n first to
AmountBy that analogy, each the matching characteristic point in the first image of correspondence, obtains n primary vector
Likewise, obtain each characteristic point in the second image as starting point, other characteristic points as terminal formed it is right
Answer multiple secondary vectors of individual features point
I.e. with i as starting point, other characteristic points j=1,2,3 ... n are terminal, form secondary vectorFor example, work as i=1
When, j is respectively 1,2,3 ... n obtains n secondary vectorAs i=2, j is respectively 1,2 ... n obtains n secondary vectorBy that analogy, each the matching characteristic point in the second image of correspondence, obtains n secondary vector
Then, the corresponding vector of each pair matching characteristic point is n primary vectorAnd n secondary vector
Step S230：Obtain each pair matching characteristic point it is corresponding vector in, each primary vector with the plurality of second to
The angle value of the angle between target secondary vector in amount, the target secondary vector are terminal and corresponding primary vector
Terminal is the secondary vector of a pair of matching characteristic points, obtains multiple angle values of correspondence each pair matching characteristic point.
For each pair matching characteristic point, its corresponding n primary vectorAnd n secondary vectorBetween have n
Individual vector angle, forms the primary vector of each angleI be equal to secondary vectorI, primary vectorJ be equal to
Secondary vectorJ.With primary vectorI and target secondary vector that the equal secondary vectors of j are the primary vector,
Such asForTarget secondary vector,ForTarget secondary vector,ForTarget secondary vector.
Calculate the angle value of the corresponding n vector angle of each pair matching characteristic point.For example, as i=1, for i-th pair
With characteristic point, primary vector is calculatedWith target secondary vectorThe angle value of the angle of formation, j are respectively 1,2,3 ... n,
Obtain n angle value.As i=2, for i-th pair matching characteristic point, primary vector is calculatedWith target secondary vector
The angle value of the angle of formation, j are respectively 1,2,3 ... n, obtains n angle value.For each pair matching characteristic point, n can be obtained
Individual angle value.
Step S240：The meansigma methodss of the corresponding multiple angle values of each pair matching characteristic point are calculated, each pair is obtained
The corresponding angular average of matching characteristic point.
The angular average of the corresponding n angle value of each pair matching characteristic point is calculated, computing formula can be：Wherein, the span of inverse cosine function is 0 to 180 degree.
In can be to obtain the corresponding angular average AverAngle of each pair matching characteristic pointi, for { (Pi, Pi') | i=
1,2, n } n obtains n angular average AverAngle to matching characteristic point, altogetheri(i=1,2,3 ... n).
Step S250：Corresponding at least one pair of the matching characteristic point of angular average for meeting default angle deletion condition is deleted
Remove, the multipair matching same place of the multipair matching characteristic point of acquisition as two width images to be matched.
In the present embodiment, the angular average of the default angle deletion condition of satisfaction can be：The institute of acquisition is angled flat
In average, predetermined ratio RaMaximum and minimum angular average, that is, meet default angle delete the maximum of condition with it is minimum
Angular average number sum be equal to all angular averages in predetermined ratio RaAngular average.Wherein, preferably, it is full
The maximum that foot deletes condition is equal with the angular average number of minimum.In the present embodiment, angular average is corresponding predetermined
Ratio RaIt is not intended as limiting, it is preferred that can be 0.15.
Specifically, calculate in n angular average, the logarithm for needing the matching characteristic point deleted is n*Ra.Again by n angle
In degree meansigma methodss, the maximum common n*R with minimumaThe corresponding matching point deletion of individual angular average.Preferably, deleteIt is individual
The maximum corresponding match point of angular average, deletesThe individual minimum corresponding match point of angular average, this enforcement
Example is mainly illustrated with this.
It should be understood that the corresponding match point of angular average is acquisition angular average AverAngleiA pair
With characteristic point, for example, the angular average of i-th pair matching characteristic point
ForOne in individual maximum angular average, then delete the i-th pair matching characteristic point.
Can be by being ranked up according to size order to n angular average, deletion comes larger
The individual maximum corresponding matching characteristic point of angular average, deletion come lessIndividual minimum angle is average
It is worth corresponding matching characteristic point.Multipair matching the same place { (P of two width images to be matched can finally be obtainediPi') | i=1,
2 ..., l }, PiFor the matching same place in the first image, Pi' it is the same place being mutually matched with Pi in the second image, can be with
Understand, l is less than n.
In embodiments of the present invention, the scheme that first embodiment and second embodiment are provided can simultaneously for obtaining
Matching characteristic point deleted, you can be after the method provided by first embodiment is deleted, then to deletion after
The matching characteristic point being mutually matched for obtaining is deleted again by the method that second embodiment is provided.Can also be logical
Cross second embodiment offer method deleted after, then the matching characteristic point being mutually matched to obtaining after deletion leads to again
The method for crossing first embodiment offer is deleted.
3rd embodiment
A kind of same place acquisition device 300 of image is present embodiments provided, as shown in figure 4, described device 300 includes：
Match point acquisition module 310, for obtaining the multipair matching characteristic being mutually matched in the first image and the second image
Point, described first image and the second image are two width images to be matched；Distance calculation module 320, for calculating described first
In image, each matching characteristic point is calculated in second image to corresponding multiple first distances of other all matching characteristic points
Each matching characteristic point is to the corresponding multiple second distances of other all matching characteristic points；Range difference computing module 330, for obtaining
Take in the corresponding distance of each pair matching characteristic point, each first distance is poor with the distance between target second distance, is corresponded to
Multiple range differences of each pair matching characteristic point, wherein, form two matching characteristic points of the target second distance and form phase
Two matching characteristic point difference Corresponding matchings of the first distance answered；Range difference mean value calculation module 340, it is described for calculating
The meansigma methodss of the corresponding multiple range differences of each pair matching characteristic point, obtain the corresponding range difference of each pair matching characteristic point average
Value；First removing module 350, for by meet predeterminable range delete condition range difference meansigma methodss it is corresponding be mutually matched
With feature point deletion, the multipair matching same place of the multipair matching characteristic point of acquisition as two width images to be matched.
In the present embodiment, meet predeterminable range delete condition range difference meansigma methodss be：The all range differences for obtaining are put down
In average, the range difference meansigma methodss of the maximum of predetermined ratio.
In addition, in the present embodiment, match point acquisition module 310 can obtain the first image and the second figure by sift algorithms
The multipair matching characteristic point being mutually matched as in.
Fourth embodiment
A kind of same place acquisition device 400 of image is present embodiments provided, as shown in figure 5, described device 400 includes：
Match point acquisition module 410, for obtaining the multipair matching characteristic being mutually matched in the first image and the second image
Point, described first image and the second image are two width images to be matched；Vectorial acquisition module 420, for first figure
Each characteristic point as in forms multiple primary vectors of each characteristic point of correspondence as starting point, other characteristic points as terminal,
Using each characteristic point in second image as starting point, other characteristic points form many of corresponding each characteristic point as terminal
Individual secondary vector；Angle calculation module 430, for obtain each pair matching characteristic point it is corresponding vector in, each primary vector with
The angle value of the angle between target secondary vector in the plurality of secondary vector, the target secondary vector are terminal and phase
The terminal of the primary vector answered is the secondary vector of a pair of matching characteristic points, obtains multiple angles of correspondence each pair matching characteristic point
Value；Angular average computing module 440, for calculating the meansigma methodss of the corresponding multiple angle values of each pair matching characteristic point,
Obtain the corresponding angular average of each pair matching characteristic point；Second removing module 450, deletes for will meet default angle
Corresponding at least one pair of matching characteristic point deletion of angular average of condition, the multipair matching characteristic point of acquisition is used as to be matched
The multipair matching same place of two width images.
Wherein, the angular average of the default angle deletion condition of satisfaction can be：In all angular averages for obtaining, in advance
The maximum of certainty ratio and minimum angular average.
In addition, match point acquisition module 410 can be obtained mutual in the first image and the second image by sift algorithms
The multipair matching characteristic point matched somebody with somebody.
It should be noted that each embodiment in this specification stress be all it is different from other embodiment it
Place, between each embodiment identical similar part mutually referring to.For device class embodiment, due to itself and method
Embodiment basic simlarity, so description is fairly simple, related part is illustrated referring to the part of embodiment of the method.
In several embodiments provided herein, it should be understood that disclosed apparatus and method, it is also possible to pass through
Other modes are realized.Device embodiment described above is only schematically, for example flow chart and block diagram in accompanying drawing
Show the device of multiple embodiments of the invention, the architectural framework in the cards of method and computer program product,
Function and operation.At this point, each square frame in flow chart or block diagram can represent the one of module, program segment or a code
Part, a part for the module, program segment or code are used to realize holding for the logic function for specifying comprising one or more
Row instruction.It should also be noted that at some as in the implementations replaced, the function of being marked in square frame can also be being different from
The order marked in accompanying drawing occurs.For example, two continuous square frames can essentially be performed substantially in parallel, and they are sometimes
Can perform in the opposite order, this is depending on involved function.It is also noted that every in block diagram and/or flow chart
The combination of individual square frame and block diagram and/or the square frame in flow chart, can use the special base for performing the function or action of regulation
Realize in the system of hardware, or can be realized with the combination of specialized hardware and computer instruction.
In addition, each functional module in each embodiment of the invention can integrate to form an independent portion
Divide, or modules individualism, it is also possible to which two or more modules are integrated to form an independent part.
If the function is realized using in the form of software function module and as independent production marketing or when using, can be with
It is stored in a computer read/write memory medium.Based on such understanding, technical scheme is substantially in other words
The part contributed to prior art or the part of the technical scheme can be embodied in the form of software product, the meter
Calculation machine software product is stored in a storage medium, is used including some instructions so that a computer equipment (can be individual
People's computer, server, or network equipment etc.) perform all or part of step of each embodiment methods described of the invention.
And aforesaid storage medium includes：USB flash disk, portable hard drive, read only memory (ROM, Read-Only Memory), random access memory are deposited
Reservoir (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with the medium of store program codes.Need
Illustrate, herein, such as first and second or the like relational terms be used merely to by an entity or operation with
Another entity or operation make a distinction, and not necessarily require or imply these entities or there is any this reality between operating
The relation or order on border.And, term " including ", "comprising" or its any other variant are intended to the bag of nonexcludability
Contain, so that a series of process, method, article or equipment including key elements is not only including those key elements, but also including
Other key elements being not expressly set out, or also include the key element intrinsic for this process, method, article or equipment.
In the absence of more restrictions, the key element for being limited by sentence "including a ...", it is not excluded that including the key element
Process, method, also there is other identical element in article or equipment.
The preferred embodiments of the present invention are the foregoing is only, the present invention is not limited to, for the skill of this area
For art personnel, the present invention can have various modifications and variations.It is all within the spirit and principles in the present invention, made any repair
Change, equivalent, improvement etc., should be included within the scope of the present invention.It should be noted that：Similar label and letter exist
Similar terms is represented in figure below, therefore, once being defined in a certain Xiang Yi accompanying drawing, then it is not required in subsequent accompanying drawing
Which is further defined and is explained.
The above, the only specific embodiment of the present invention, but protection scope of the present invention is not limited thereto, any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all be contained
Cover within protection scope of the present invention.Therefore, protection scope of the present invention described should be defined by scope of the claims.
Claims (10)
1. same place acquisition methods of a kind of image, it is characterised in that methods described includes：
The multipair matching characteristic point being mutually matched in obtaining the first image and the second image, described first image with the second image is
Two width images to be matched；
In calculating described first image, each matching characteristic point is counted to corresponding multiple first distances of other all matching characteristic points
Each matching characteristic point is calculated in second image to the corresponding multiple second distances of other all matching characteristic points；
Obtain in the corresponding distance of each pair matching characteristic point, each first distance is poor with the distance between target second distance, obtains
Multiple range differences of each pair matching characteristic point must be corresponded to, wherein, formed two matching characteristic points of the target second distance with
Form two matching characteristic point difference Corresponding matchings of corresponding first distance；
The meansigma methodss of the corresponding multiple range differences of each pair matching characteristic point are calculated, each pair matching characteristic point correspondence is obtained
Range difference meansigma methodss；
The corresponding matching characteristic point deletion being mutually matched of range difference meansigma methodss of predeterminable range deletion condition will be met, acquisition
Multipair matching same place of the multipair matching characteristic point as two width images to be matched.
2. method according to claim 1, it is characterised in that it is described meet predeterminable range delete condition range difference it is average
It is worth and is：In all range difference meansigma methodss for obtaining, the range difference meansigma methodss of the maximum of predetermined ratio.
3. method according to claim 1, it is characterised in that be mutually matched in the first image of the acquisition and the second image
Multipair matching characteristic point include：
The multipair matching characteristic point being mutually matched in first image and the second image is obtained by sift algorithms.
4. same place acquisition methods of a kind of image, it is characterised in that methods described includes：
The multipair matching characteristic point being mutually matched in obtaining the first image and the second image, described first image with the second image is
Two width images to be matched；
Using each characteristic point in described first image as starting point, other characteristic points form corresponding each characteristic point as terminal
Multiple primary vectors, using each characteristic point in second image as starting point, other characteristic points form right as terminal
Answer multiple secondary vectors of each characteristic point；
Obtain in the corresponding vector of each pair matching characteristic point, the target second in each primary vector and the plurality of secondary vector
The angle value of the angle between vector, the target secondary vector are that terminal is matched for a pair with the terminal of corresponding primary vector
The secondary vector of characteristic point, obtains multiple angle values of correspondence each pair matching characteristic point；
The meansigma methodss of the corresponding multiple angle values of each pair matching characteristic point are calculated, each pair matching characteristic point correspondence is obtained
Angular average；
Default angle will be met delete corresponding at least one pair of matching characteristic point deletion of angular average of condition, acquisition it is multipair
Multipair matching same place of the matching characteristic point as two width images to be matched.
5. method according to claim 4, it is characterised in that the default angle of the satisfaction deletes the angular average of condition
For：In all angular averages for obtaining, the angular average of the maximum and minimum of predetermined ratio.
6. method according to claim 4, it is characterised in that be mutually matched in the first image of the acquisition and the second image
Multipair matching characteristic point include：
The multipair matching characteristic point being mutually matched in first image and the second image is obtained by sift algorithms.
7. the same place acquisition device of a kind of image, it is characterised in that described device includes：
Match point acquisition module, it is for obtaining the multipair matching characteristic point being mutually matched in the first image and the second image, described
First image and the second image are two width images to be matched；
Distance calculation module, it is corresponding to other all matching characteristic points for calculating each matching characteristic point in described first image
Multiple first distances, in calculating second image, each matching characteristic point is corresponding multiple to other all matching characteristic points
Second distance；
Range difference computing module, for obtaining in the corresponding distance of each pair matching characteristic point, each first distance and target second
The distance between distance is poor, obtains multiple range differences of correspondence each pair matching characteristic point, wherein, form the target second distance
Two matching characteristic points the first distance corresponding with formation two matching characteristic points difference Corresponding matchings；
Range difference mean value calculation module, for calculating the meansigma methodss of the corresponding multiple range differences of each pair matching characteristic point,
Obtain the corresponding range difference meansigma methodss of each pair matching characteristic point；
First removing module, for the corresponding matching being mutually matched of range difference meansigma methodss by predeterminable range deletion condition is met
Feature point deletion, the multipair matching same place of the multipair matching characteristic point of acquisition as two width images to be matched.
8. device according to claim 7, it is characterised in that it is described meet predeterminable range delete condition range difference it is average
It is worth and is：In all range difference meansigma methodss for obtaining, the range difference meansigma methodss of the maximum of predetermined ratio.
9. the same place acquisition device of a kind of image, it is characterised in that described device includes：
Match point acquisition module, it is for obtaining the multipair matching characteristic point being mutually matched in the first image and the second image, described
First image and the second image are two width images to be matched；
Vectorial acquisition module, for using each characteristic point in described first image as starting point, other characteristic points are used as terminal
Formation corresponds to multiple primary vectors of each characteristic point, and using each characteristic point in second image as starting point, other are special
Levy multiple secondary vectors that each characteristic point of correspondence is a little formed as terminal；
Angle calculation module, for obtaining in the corresponding vector of each pair matching characteristic point, each primary vector and the plurality of the
The angle value of the angle between target secondary vector in two vectors, the target secondary vector be terminal with corresponding first to
The terminal of amount is the secondary vector of a pair of matching characteristic points, obtains multiple angle values of correspondence each pair matching characteristic point；
Angular average computing module, for calculating the meansigma methodss of the corresponding multiple angle values of each pair matching characteristic point, obtains
Obtain the corresponding angular average of each pair matching characteristic point；
Second removing module, for corresponding at least one pair of matching characteristic of angular average by default angle deletion condition is met
Point deletion, the multipair matching same place of the multipair matching characteristic point of acquisition as two width images to be matched.
10. device according to claim 9, it is characterised in that the angle that the default angle of the satisfaction deletes condition is average
It is worth and is：In all angular averages for obtaining, the angular average of the maximum and minimum of predetermined ratio.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611125944.7A CN106529607B (en) | 2016-12-09 | 2016-12-09 | The same place acquisition methods and device of image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611125944.7A CN106529607B (en) | 2016-12-09 | 2016-12-09 | The same place acquisition methods and device of image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106529607A true CN106529607A (en) | 2017-03-22 |
| CN106529607B CN106529607B (en) | 2019-06-18 |
Family
ID=58342425
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611125944.7A Active CN106529607B (en) | 2016-12-09 | 2016-12-09 | The same place acquisition methods and device of image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106529607B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113591977A (en) * | 2021-07-29 | 2021-11-02 | 武汉联影智融医疗科技有限公司 | Point pair matching method and device, electronic equipment and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010020624A1 (en) * | 2008-08-20 | 2010-02-25 | European Aeronautic Defence And Space Company - Eads France | Method of metric telemetry |
| CN103236064A (en) * | 2013-05-06 | 2013-08-07 | 东南大学 | Point cloud automatic registration method based on normal vector |
| CN104331626A (en) * | 2014-11-08 | 2015-02-04 | 北京握奇智能科技有限公司 | Map matching method and map matching system based on floating car satellite positioning data |
| CN104700451A (en) * | 2015-03-14 | 2015-06-10 | 西安电子科技大学 | Point cloud registering method based on iterative closest point algorithm |
| CN105184801A (en) * | 2015-09-28 | 2015-12-23 | 武汉大学 | Optical and SAR image high-precision registration method based on multilevel strategy |
| WO2015166087A3 (en) * | 2014-04-30 | 2016-02-18 | Piksel, Inc | Device synchronization |
-
2016
- 2016-12-09 CN CN201611125944.7A patent/CN106529607B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010020624A1 (en) * | 2008-08-20 | 2010-02-25 | European Aeronautic Defence And Space Company - Eads France | Method of metric telemetry |
| CN103236064A (en) * | 2013-05-06 | 2013-08-07 | 东南大学 | Point cloud automatic registration method based on normal vector |
| WO2015166087A3 (en) * | 2014-04-30 | 2016-02-18 | Piksel, Inc | Device synchronization |
| CN104331626A (en) * | 2014-11-08 | 2015-02-04 | 北京握奇智能科技有限公司 | Map matching method and map matching system based on floating car satellite positioning data |
| CN104700451A (en) * | 2015-03-14 | 2015-06-10 | 西安电子科技大学 | Point cloud registering method based on iterative closest point algorithm |
| CN105184801A (en) * | 2015-09-28 | 2015-12-23 | 武汉大学 | Optical and SAR image high-precision registration method based on multilevel strategy |
Non-Patent Citations (7)
| Title |
|---|
| CHUI, HL 等: "A new point matching algorithm for non-rigid registration", 《COMPUTER VISION AND IMAGE UNDERSTANDING》 * |
| TAM, GARY K. L. 等: "Registration of 3D Point Clouds and Meshes: A Survey from Rigid to Nonrigid", 《IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS》 * |
| 佘毅 等: "利用薄板样条函数的无人机飞行质量检查", 《遥感信息》 * |
| 佘毅 等: "高速铁路轨道车载近景影像的精确匹配与建模", 《铁道工程学报》 * |
| 熊文成 等: "环境一号卫星A/B星数据自动几何精校正设计研究", 《中国工程科学》 * |
| 王蕊 等: "基于几何特征的点云配准算法", 《华东理工大学学报》 * |
| 靳峰 等: "结合图像信息和几何关系的配准算法", 《华中科技大学学报 》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113591977A (en) * | 2021-07-29 | 2021-11-02 | 武汉联影智融医疗科技有限公司 | Point pair matching method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106529607B (en) | 2019-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109376631B (en) | Loop detection method and device based on neural network | |
| CN110826519A (en) | Face occlusion detection method and device, computer equipment and storage medium | |
| CN106778587B (en) | State of flight detection method and device based on unmanned plane image | |
| CN107240078A (en) | Lens articulation Method for Checking, device and electronic equipment | |
| US10289884B2 (en) | Image analyzer, image analysis method, computer program product, and image analysis system | |
| CN110298858B (en) | Image clipping method and device | |
| CN108550166B (en) | Spatial target image matching method | |
| CN109215131B (en) | Virtual face driving method and device | |
| CN113111844B (en) | Operation posture evaluation method and device, local terminal and readable storage medium | |
| CN105261031B (en) | The line detection method and device calculated based on gradient | |
| US20160379088A1 (en) | Apparatus and method for creating an image recognizing program having high positional recognition accuracy | |
| Kumar et al. | Adaptive cluster tendency visualization and anomaly detection for streaming data | |
| Wang et al. | Fast and robust object tracking via probability continuous outlier model | |
| CN108829804A (en) | Based on the high dimensional data similarity join querying method and device apart from partition tree | |
| US20200005078A1 (en) | Content aware forensic detection of image manipulations | |
| CN113420848A (en) | Neural network model training method and device and gesture recognition method and device | |
| CN110689230A (en) | Method for determining poverty degree of region, electronic device and storage medium | |
| CN112614074B (en) | Robust vanishing point detection method and device based on response graph and clustering | |
| CN111680183B (en) | Object retrieval method and device, storage medium and electronic equipment | |
| CN113313217A (en) | Method and system for accurately identifying dip angle characters based on robust template | |
| CN106529607A (en) | Method and device for acquiring homonymy points of images | |
| CN109451318B (en) | Method, apparatus, electronic device and storage medium for facilitating VR video encoding | |
| CN110956131A (en) | Single-target tracking method, device and system | |
| CN110766938A (en) | Road network topological structure construction method and device, computer equipment and storage medium | |
| CN112699863B (en) | Fingerprint enhancement method, computer-readable storage medium and electronic device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: No.7 Jiuxing Avenue, high tech Zone, Chengdu, Sichuan 610000 Patentee after: Sichuan Surveying and Mapping Product Quality Supervision and Inspection Station, Ministry of Natural Resources (Sichuan Surveying and Mapping Product Quality Supervision and Inspection Station) Address before: No.7 Jiuxing Avenue, high tech Zone, Chengdu, Sichuan 610000 Patentee before: SICHUAN MAPPING PRODUCT QUALITY SUPERVISION AND INSPECTION STATION, NATIONAL ADMINISTRATION OF SURVEYING, MAPPING AND GEOINFORMATION |