CN108983769A - Immediately the optimization method and device of positioning and map structuring - Google Patents
Immediately the optimization method and device of positioning and map structuring Download PDFInfo
- Publication number
- CN108983769A CN108983769A CN201810650981.2A CN201810650981A CN108983769A CN 108983769 A CN108983769 A CN 108983769A CN 201810650981 A CN201810650981 A CN 201810650981A CN 108983769 A CN108983769 A CN 108983769A
- Authority
- CN
- China
- Prior art keywords
- map
- stability features
- image
- processed
- stability
- 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
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
- G05D1/0251—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means extracting 3D information from a plurality of images taken from different locations, e.g. stereo vision
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Abstract
The present invention provides the optimization method and devices of a kind of positioning and map structuring immediately.The instant positioning and the optimization method of map structuring include: the stability features for obtaining image to be processed；The similar stability features are shielded, to construct initial map and position according to the stability features after shielding；The image to be processed according to the stability features, after tracking initialization process；According to the stability features, processing is optimized to the map of building and position.The technical program obtains the stability features of image to be processed, it on the one hand can be according to the similar stability features of shielding, to map and position carry out initialization process, it on the other hand can be according to the stability features, track and optimize the image to be processed, it is more stable since stability features are obtained according to the attribute of object, therefore the loss of feature can be effectively prevented in map and the initialization of position benefit, tracking and optimization process in this way.
Description
Technical field
The present invention relates to technical field of computer vision, in particular to a kind of optimization method of positioning and map structuring immediately
And device.
Background technique
Immediately positioning and map structuring (simultaneous localization and mapping, SLAM), also referred to as
CML (Concurrent Mapping and Localization), refer to robot under the conditions of self-position is uncertain,
Map is created in complete graphics communication, while utilizing map autonomous positioning and navigation.View-based access control model sensor positioning method is
The hot spot studied both at home and abroad in recent years, it is divided into monocular, binocular and the positioning of more mesh again.Currently, for monocular SLAM method
Research focus on indoors or under road scene, improve positioning and the speed and accuracy of composition.In these scenes, feature
Major part is differentiable, therefore is easy to estimate camera posture according to matched feature or pixel.However, in many scenes
In, the problem of being not easy to distinguishing characteristic, will lead to scale drift and pursuing missing in this way.
In traditional technology, solution to this problem has indirect method (ORB-SLAM), direct method (lsd-SLAM) etc..Its
In, indirect method is all made very well in most of scenes, and contains the complete function including relocating, but it is very
Feature extraction and characteristic matching are depended in big degree.Direct method directly uses actual sensor values, but in large scene very
The problem of being difficult to resolve certainly scale drift and relocating.
Summary of the invention
The present invention provides the optimization method and device of positioning and map structuring immediately, to solve to propose in background technique
One or more technical problems provide a kind of beneficial selection.
As one aspect of the present invention, the embodiment of the present invention provides the optimization side of a kind of positioning and map structuring immediately
Method, comprising:
Obtain the stability features of image to be processed；
The similar stability features are shielded, to construct initial map and position according to the stability features；
The image to be processed according to the stability features, after tracking initialization process；
According to the stability features, processing is optimized to the map of building and position.
With reference to first aspect, the embodiment of the present invention is in the first embodiment of first aspect, the stability features packet
Include following at least one feature: angle point, shape, size, color.
First embodiment with reference to first aspect, the embodiment of the present invention is in the second embodiment of first aspect, screen
The similar stability features are covered, to carry out initialization process for map and position, comprising:
Obtain the subcharacter of the stability features；
The subcharacter is subjected to clustering processing, to obtain the subcharacter that wherein similarity is greater than preset threshold；
Shield the subcharacter that the similarity is greater than preset threshold.
First embodiment with reference to first aspect, the embodiment of the present invention is in the third embodiment of first aspect, screen
The similar stability features are covered, to carry out initialization process for map and position, comprising:
Divide the similar stability characteristic area of the image to be processed using default machine learning model, described in shielding
Similar stability characteristic area；Wherein, the similarity of the stability features in the similar stability characteristic area is greater than default
Threshold value.
With reference to first aspect, the embodiment of the present invention is special according to the stability in the 4th embodiment of first aspect
Sign, optimizes processing to the map of building and position, comprising:
According to the tracking situation of object indicated by the image to be processed, corrected using corresponding modification method described
Map and the drift of the scale of position.
With reference to first aspect, the embodiment of the present invention is in the first implementation of the 4th embodiment of first aspect,
According to the tracking situation of object indicated by the image to be processed, the map and position are corrected using corresponding modification method
The scale drift set, comprising:
If tracking object indicated by the image to be processed can be tracked, corrected describedly using light-stream adjustment
Figure and the scale of position drift about.
The 4th embodiment with reference to first aspect, in the embodiment of the present invention in second of realization side of the 4th embodiment
In formula, according to the tracking situation of object indicated by the image to be processed, corrected describedly using corresponding modification method
Figure and the scale of position drift about, comprising:
If object indicated by the image to be processed can not be tracked, the map is corrected by default increasing function
It drifts about with the scale of position.
Second aspect, the embodiment of the present invention provide a kind of position immediately and include: with map structuring device, described device
Module is obtained, is configured to obtain the stability features of image to be processed；
Shroud module is configured to shield the similar stability features, to carry out at initialization for map and position
Reason；
Tracing module is configured to the image to be processed according to the stability features, after tracking initialization process；
Optimization module, is configured to according to the stability features, optimizes processing to the map of building and position.
In conjunction with second aspect, in the first embodiment of the second aspect of the embodiment of the present invention, the shroud module includes:
Acquisition submodule is configured to obtain the subcharacter of the stability features；
Submodule is clustered, is configured to the subcharacter carrying out clustering processing, is greater than default threshold to obtain wherein similarity
The subcharacter of value；
First shielding submodule is configured to shield the subcharacter that the similarity is greater than preset threshold.
In conjunction with second aspect, in the second embodiment of the second aspect of the embodiment of the present invention, the initialization module packet
It includes:
Secondary shielding submodule is configured to divide the similar stable of the image to be processed using default machine learning model
Property characteristic area, to shield the similar stability characteristic area；Wherein, the stability in the similar stability characteristic area
The similarity of feature is greater than preset threshold.
The present invention by adopting the above technical scheme, has the advantages that the technical program obtains the stability of image to be processed
Feature, on the one hand can be according to the similar stability features be shielded, and to map and position carry out initialization process, another party
The image to be processed can be tracked and be optimized in face, since stability features are according to object according to the stability features
What attribute obtained, it is more stable, therefore can be effectively prevented in map and the initialization of position benefit, tracking and optimization process in this way
The loss of feature.
Detailed description of the invention
In the accompanying drawings, unless specified otherwise herein, otherwise indicate the same or similar through the identical appended drawing reference of multiple attached drawings
Component or element.What these attached drawings were not necessarily to scale.It should be understood that these attached drawings depict only according to the present invention
Disclosed some embodiments, and should not serve to limit the scope of the present invention.
Fig. 1 is the flow chart of the instant positioning and the optimization method of map structuring of the embodiment of the present invention one；
Fig. 2 is the flow chart of the instant positioning and the optimization method of map structuring of the embodiment of the present invention two；
Fig. 3 (1) be the embodiment of the present invention two instant positioning in the optimization method of map structuring to image to be processed into
The image of solar panel before row phase region segmentation of feature regions；
Fig. 3 (2) is in the instant positioning of the embodiment of the present invention two and the optimization method of map structuring by solar panel
Image after similar features regions shield；
Fig. 4 is the signal of the instant positioning and generation scale drift in the optimization method of map structuring of the embodiment of the present invention two
Figure；
Fig. 5 is that the instant positioning of the embodiment of the present invention two is corrected with the optimization method of map structuring according to the attribute of object
The schematic illustration of scale drift；
Fig. 6 is the ruler that image to be processed is corrected in the instant positioning of the embodiment of the present invention two and the optimization method of map structuring
Spend the schematic diagram of drift；
Fig. 7 is the ruler that image to be processed is corrected in the instant positioning of the embodiment of the present invention two and the optimization method of map structuring
Spend the schematic diagram of drift；
Fig. 8 is the schematic diagram of the instant positioning and map structuring device of the embodiment of the present invention three.
Specific embodiment
Hereinafter, certain exemplary embodiments are simply just described.As one skilled in the art will recognize that
Like that, without departing from the spirit or scope of the present invention, described embodiment can be modified by various different modes.
Therefore, attached drawing and description are considered essentially illustrative rather than restrictive.
In the description of the present invention, it is to be understood that, term " center ", " longitudinal direction ", " transverse direction ", " length ", " width ",
" thickness ", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outside", " up time
The orientation or positional relationship of the instructions such as needle ", " counterclockwise ", " axial direction ", " radial direction ", " circumferential direction " be orientation based on the figure or
Positional relationship is merely for convenience of description of the present invention and simplification of the description, rather than the device or element of indication or suggestion meaning must
There must be specific orientation, be constructed and operated in a specific orientation, therefore be not considered as limiting the invention.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include one or more of the features.In the description of the present invention, the meaning of " plurality " is two or more,
Unless otherwise specifically defined.
In the present invention unless specifically defined or limited otherwise, term " installation ", " connected ", " connection ", " fixation " etc.
Term shall be understood in a broad sense, for example, it may be being fixedly connected, may be a detachable connection, or integral；It can be mechanical connect
It connects, is also possible to be electrically connected, can also be communication；It can be directly connected, can also indirectly connected through an intermediary, it can be with
It is the interaction relationship of the connection or two elements inside two elements.For the ordinary skill in the art, may be used
To understand the concrete meaning of above-mentioned term in the present invention as the case may be.
In the present invention unless specifically defined or limited otherwise, fisrt feature second feature "upper" or "lower"
It may include that the first and second features directly contact, also may include that the first and second features are not direct contacts but pass through it
Between other characterisation contact.Moreover, fisrt feature second feature " on ", " side " and " above " include fisrt feature
Right above second feature and oblique upper, or first feature horizontal height is merely representative of higher than second feature.Fisrt feature is
Two features " under ", " lower section " and " following " include fisrt feature right above second feature and oblique upper, or be merely representative of
One characteristic level height is less than second feature.
Following disclosure provides many different embodiments or example is used to realize different structure of the invention.In order to
Simplify disclosure of the invention, hereinafter the component of specific examples and setting are described.Certainly, they are merely examples, and
And it is not intended to limit the present invention.In addition, the present invention can in different examples repeat reference numerals and/or reference letter,
This repetition is for purposes of simplicity and clarity, itself not indicate between discussed various embodiments and/or setting
Relationship.In addition, the present invention provides various specific techniques and material example, but those of ordinary skill in the art can be with
Recognize the application of other techniques and/or the use of other materials.
Currently, indirect method is widely used, but its performance is affected by presence feature.For example, in repeated characteristic field
Jing Zhong, due to the influence of similar characteristic, indirect method is difficult to initialize and be easily lost.
Embodiment one
As shown in Figure 1, the schematic diagram of instant positioning and the optimization method of map structuring for the embodiment of the present invention one.This hair
The instant positioning of bright embodiment one and the optimization method of map structuring include:
S101 obtains the stability features of image to be processed.
The image to be processed of the embodiment of the present invention can be obtained by the camera being installed on smart machine.For example, intelligence
Robot.
In the specific implementation, since map and position are three-dimensional space, and image belongs to two dimensional image to the embodiment of the present invention,
Therefore the embodiment of the present invention obtains at least two images or picture frame to be processed generally to handle.Firstly the need of acquisition figure to be processed
As upper stability features.
Wherein, the stability features comprise at least one of the following feature: angle point, shape, size, color.
Stability features are obtained according to the attribute of object.For example, stability features can be angle point or some
Point on special construction.So-called angle point is exactly extreme point, i.e., the particularly pertinent point of attribute in terms of certain.It is of course also possible to according to
The actual demand of user oneself to define the attribute (specific entropy is arranged and carries out Corner Detection) of angle point.Angle point can be two
The infall of line is also possible to the point being located in the different things of two adjacent Main ways.Point on some special constructions,
Refer to the point that thingness on special object, can be identified, for example, the profile or shape of graphical object.
S102 shields the similar stability features, to be constructed initially according to the stability features after shielding
Figure and position.
It, can if there are similar features for two images to be processed during to map and position are initialized
The calculating of the characteristic matching and fundamental matrix or homography matrix in initialization procedure is influenced, therefore the embodiment of the present invention is right first
In masking similar features, to avoid repeated characteristic.Here not by all, there are the features of similitude all to mask, but
Retain at least one or several feature therein, other Feature maskings similar with this one or several feature are fallen.
Those skilled in the art should be it is recognised that so-called to map and position progress initialization process, be according to extremely
Corresponding characteristic point in few two images to be processed, to establish three-dimensional map, and positions object.
The feature on image to be processed can be compared by different methods, for example, can also lead to by the method for cluster
Method of over-segmentation image similar area etc. determines similar features.
S103, the image to be processed according to the stability features, after tracking initialization process.
Since stability features are obtained according to the attribute of object, there is stability.Therefore it may be referred to object
Stability features track the image to be processed.
S104 optimizes processing to the map of building and position according to the stability features.
Similarly, since stability features are obtained according to the attribute of object, there is stability.Therefore, Ke Yican
The stability features of object are examined to optimize the image to be processed.
It is worth noting that, the embodiment of the present invention is not actually in the stability features step for obtaining image to be processed
While only obtaining the stability features of image to be processed, but obtain stability features, the base of image to be processed is also obtained
Eigen.When shielding similar features, while essential characteristic and stability features being shielded, is subsequently used for map and position
Initialization.The embodiment of the present invention is in order to succinct when introducing the technical program, without focusing on pen and ink introduction for substantially special
The shielding processing of sign.But it will be recognized by those skilled in the art that being to all features, i.e., when obtaining feature and shielding characteristic
Essential characteristic and stability features, while being handled.
The technical program obtains the stability features of image to be processed, on the one hand can be according to the similar stabilization of shielding
Property feature, to map and position carry out initialization process, on the other hand can be tracked and optimization institute according to the stability features
Image to be processed is stated, it is more stable since stability features are obtained according to the attribute of object, therefore in this way in map and position
Set the loss that feature can be effectively prevented in benefit initialization, tracking and optimization process.
Embodiment two
On the basis of example 1, the embodiment of the present invention two introduces realization process of the invention in more detail.Such as Fig. 2
It is shown, it is the flow chart of the instant positioning and the optimization method of map structuring of the embodiment of the present invention two.The embodiment of the present invention one
Immediately positioning and the optimization method of map structuring include:
S201 obtains the stability features of image to be processed.
The step S101 of the step S201 corresponding embodiment one of the present embodiment.
S202 obtains the subcharacter of the stability features.
The present embodiments relate to subcharacter, not for the division again of single feature, but by all features
It is divided into different subclass.Since once all features can not be handled simultaneously, it, can be by after being divided into subclass
A processing improves processing speed.
S203 carries out clustering processing to the subcharacter, to obtain the subcharacter that wherein similarity is greater than preset threshold.
So-called cluster refers to the process of that the set by physics or abstract object is divided into the multiple classes being made of similar object
Referred to as cluster.By clustering the set that cluster generated is one group of data object, object in these objects and the same cluster that
This is similar, different with the object in other clusters.
The embodiment of the present invention carries out clustering processing to subcharacter, can determine wherein similar subcharacter.
In other embodiments of the invention, it is special that the similar stability of shielding can also be realized by the following method
Sign:
Divide the similar stability characteristic area of the image to be processed using default machine learning model, described in shielding
Similar stability characteristic area；Wherein, the similarity of the characteristic point in the similar stability characteristic area is greater than preset threshold.
S204 shields the subcharacter that the similarity is greater than preset threshold.
The embodiment of the present invention can train default machine learning model using machine learning algorithm, make the machine learning mould
Type can filter or detect the similar features region in image to be processed.As shown in Fig. 3 (1), for the solar energy face before segmentation
The image of plate, Fig. 3 (2) are by the image after solar panel similar features regions shield.
The step S202 of the present embodiment to S204 corresponding embodiment one step S102.
S205 tracks the image to be processed according to the stability features.
The step S103 of the step S204 corresponding embodiment one of the present embodiment.
In general, smart machine is to be positioned by image procossing and track object.In tracing process, when to be processed
When being full of similar features on image, it is easy to cause pursuing missing.When using traditional technology, the spy that is detected on time k
Point is levied, is surely detected time k+1 is different.To prevent pursuing missing, the embodiment of the present invention utilizes the attribute of object, Ke Yishi
Present different time can extract identical characteristic point.For example, using the angle point of object as stability features.It is tracking in this way
When object, each time can get the angle point of the object, prevent tracked missing image.
The method for wherein tracking the figure to be processed can be with reference to the method for tracing in traditional technology, and details are not described herein.
S206 optimizes processing to the map of building and position according to the stability features.
Wherein, scale drift is the intrinsic problem of monocular SLAM.As shown in figure 4, same when being observed on different time
When object, since scale drift occurs, the object may be considered being different object on other visual angles.Moreover, once
Scale drift occurs, mistake will occur in the map of subsequent builds and position.As shown in figure 5, when camera is to three rectangle objects
Body, i.e. object 1, object 2 and object 3, when photograph, since the distance of three rectangle objects is different, image hair that camera is formed
Raw overlapping.At this moment, can be according to the size characteristic in the stability features of these three rectangle objects, it can be according to size characteristic pair
The map currently constructed carries out scale amendment.Wherein, size is one of attribute of object, therefore size characteristic is the category by object
Property can obtain.
Wherein, in the scale drift for correcting the map and position, following two situation can be divided into and handled respectively.
If tracking object indicated by the image to be processed can be tracked, corrected describedly using light-stream adjustment
Figure and the scale of position drift about.
As shown in fig. 6, observing the same object respectively in time k, k+i and k+j.In each time, the object is all
It may map in three-dimensional coordinate.Since scale drifts about, position and size of the object in three-dimensional coordinate are all different, therefore
An energy function can be defined to describe this species diversity:
Wherein, riAnd rjIt is the object observed in time i and j, function S (ri, rj) indicate in time riWith time rjWhen,
Shape or difference in size of the object in three-dimensional coordinate.
Since estimation of the object in three-dimensional coordinate is related with the posture of camera, optimization camera can be passed through
Posture minimizes position and difference in size of the object in different time.Calculation formula is as follows:
Wherein, T indicates the posture of camera, EerrIndicate position and difference in size of the object in different time,Indicating the key frame set for needing to optimize, R indicates the object observed in key frame set,
If object indicated by the image to be processed can not be tracked, the map is corrected by default increasing function
It drifts about with the scale of position.
When object cannot be tracked, then position and the size of different time object can not be obtained, this is just needed to each
New key frame carries out size estimation.As shown in fig. 7, observing object in key frame, and obtain its position on three-dimensional coordinate
It sets.If difference between known actual shape can be with is defined as:
Dr(rk)=αkD(rk)+v (3)
Wherein αkIndicate that amendment scale parameter, v indicate that wrong noise, k indicate key frame, D (rk) indicate shape estimation.
When a new key frame is inserted into image to be processed, the camera posture of this frame can be passed by following
Increasing function obtains:
Wherein,Indicate camera posture after correcting, TkT-1 k-1It indicates one coordinate in k-1 frame is transformed into k frame
Transition matrix, functionIt indicates similarity transformation function, can be used for updating the posture of camera, αkIt indicates to need
The stochastic variable estimated in kth frame.
The technical program is handled with reference to stability features in tracking and Optimization Steps；It is on the one hand chasing after in this way
During track, system is allow steadily to estimate the posture of camera；It on the other hand, can be according to the shape of object in optimization process
Shape or size are drifted about come scale when correcting building map.
Embodiment three
The embodiment of the present invention three provides a kind of positioning and map structuring device immediately.As shown in figure 8, the embodiment of the present invention
Instant positioning with map structuring device include:
Module 81 is obtained, is configured to obtain the stability features of image to be processed；
Shroud module 82 is configured to shield the similar stability features, to be constructed just according to the stability features
Beginning map and position；
Tracing module 83 is configured to the image to be processed according to the stability features, after tracking initialization process；
Optimization module 84, is configured to according to the stability features, optimizes place to the map of building and position
Reason.
Further, the shroud module includes:
Acquisition submodule 821 is configured to obtain the subcharacter of the stability features；
Cluster submodule 822, be configured to the subcharacter carrying out clustering processing, with obtain wherein similarity be greater than it is default
The subcharacter of threshold value；
First shielding submodule 823 is configured to shield the subcharacter that the similarity is greater than preset threshold.
In another alternate embodiment of the invention, the shroud module 82 can also include:
Secondary shielding submodule (not shown) is configured to divide the figure to be processed using default machine learning model
The similar stability characteristic area of picture, to shield the similar stability characteristic area；Wherein, the similar stability characteristic area
The similarity of stability features in domain is greater than preset threshold.
The technical program has the advantages that the technical program by referring in initialization, tracking and optimization process
The stability features of image to be processed, to prevent from losing feature in tracing process.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can readily occur in its various change or replacement,
These should be covered by the protection scope of the present invention.Therefore, protection scope of the present invention should be with the guarantor of the claim
It protects subject to range.
Claims (10)
1. a kind of optimization method of positioning and map structuring immediately, which is characterized in that the described method includes:
Obtain the stability features of image to be processed；
The similar stability features are shielded, to construct initial map and position according to the stability features after shielding；
According to the stability features, the image to be processed is tracked；
According to the stability features, row optimization processing is carried out to the map of building and position.
2. the method according to claim 1, wherein the stability features comprise at least one of the following feature:
Angle point, shape, size, color.
3. the method according to claim 1, wherein the similar stability features are shielded, to be used for map
Initialization process is carried out with position, comprising:
Obtain the subcharacter of the stability features；
The subcharacter is subjected to clustering processing, to obtain the subcharacter that wherein similarity is greater than preset threshold；
Shield the subcharacter that the similarity is greater than preset threshold.
4. the method according to claim 1, wherein the similar stability features are shielded, to be used for map
Initialization process is carried out with position, comprising:
Divide the similar stability characteristic area of the image to be processed using default machine learning model, it is described similar to shield
Stability features region；Wherein, the similarity of the stability features in the similar stability characteristic area is greater than preset threshold.
5. the method according to claim 1, wherein according to the stability features, to the map of building
Processing is optimized with position, comprising:
According to the tracking situation of object indicated by the image to be processed, the map is corrected using corresponding modification method
It drifts about with the scale of position.
6. according to the method described in claim 5, it is characterized in that, the tracking of the object according to indicated by the image to be processed
Situation corrects the scale drift of the map and position using corresponding modification method, comprising:
If object indicated by the tracking image to be processed can be tracked, corrected using light-stream adjustment the map with
The scale of position drifts about.
7. according to the method described in claim 5, it is characterized in that, the tracking of the object according to indicated by the image to be processed
Situation corrects the scale drift of the map and position using corresponding modification method, comprising:
If object indicated by the image to be processed can not be tracked, the map and position are corrected by default increasing function
The scale drift set.
8. a kind of positioning immediately and map structuring device, which is characterized in that described device includes:
Module is obtained, is configured to obtain the stability features of image to be processed；
Shroud module is configured to shield the similar stability features, to carry out initialization process for map and position；
Tracing module is configured to the image to be processed according to the stability features, after tracking initialization process；
Optimization module, is configured to according to the stability features, optimizes processing to the map of building and position.
9. device according to claim 8, which is characterized in that the shroud module includes:
Acquisition submodule is configured to obtain the subcharacter of the stability features；
Submodule is clustered, is configured to the subcharacter carrying out clustering processing, is greater than preset threshold to obtain wherein similarity
Subcharacter；
First shielding submodule is configured to shield the subcharacter that the similarity is greater than preset threshold.
10. device according to claim 8, which is characterized in that the shroud module includes:
Secondary shielding submodule, the similar stability for being configured to divide using default machine learning model the image to be processed are special
Region is levied, to shield the similar stability characteristic area；Wherein, the stability features in the similar stability characteristic area
Similarity be greater than preset threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810650981.2A CN108983769B (en) | 2018-06-22 | 2018-06-22 | Instant positioning and map construction optimization method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810650981.2A CN108983769B (en) | 2018-06-22 | 2018-06-22 | Instant positioning and map construction optimization method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108983769A true CN108983769A (en) | 2018-12-11 |
| CN108983769B CN108983769B (en) | 2022-06-21 |
Family
ID=64538147
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810650981.2A Active CN108983769B (en) | 2018-06-22 | 2018-06-22 | Instant positioning and map construction optimization method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108983769B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109887003A (en) * | 2019-01-23 | 2019-06-14 | 亮风台（上海）信息科技有限公司 | A kind of method and apparatus initialized for carrying out three-dimensional tracking |
| CN109965797A (en) * | 2019-03-07 | 2019-07-05 | 深圳市愚公科技有限公司 | Generation method, sweeping robot control method and the terminal of sweeping robot map |
| CN112767818A (en) * | 2019-11-01 | 2021-05-07 | 北京初速度科技有限公司 | Map construction method and device |
| CN113220818A (en) * | 2021-05-27 | 2021-08-06 | 南昌智能新能源汽车研究院 | Automatic mapping and high-precision positioning method for parking lot |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103886107A (en) * | 2014-04-14 | 2014-06-25 | 苏州市华天雄信息科技有限公司 | Robot locating and map building system based on ceiling image information |
| CN103900583A (en) * | 2012-12-25 | 2014-07-02 | 联想(北京)有限公司 | Device and method used for real-time positioning and map building |
| CN104574421A (en) * | 2015-01-29 | 2015-04-29 | 北方工业大学 | Large-breadth small-overlapping-area high-precision multispectral image registration method and device |
| CN105825520A (en) * | 2015-01-08 | 2016-08-03 | 北京雷动云合智能技术有限公司 | Monocular SLAM (Simultaneous Localization and Mapping) method capable of creating large-scale map |
| CN106355197A (en) * | 2016-08-24 | 2017-01-25 | 广东宝乐机器人股份有限公司 | Navigation image matching filtering method based on K-means clustering algorithm |
| WO2018089703A1 (en) * | 2016-11-09 | 2018-05-17 | The Texas A&M University System | Method and system for accurate long term simultaneous localization and mapping with absolute orientation sensing |
-
2018
- 2018-06-22 CN CN201810650981.2A patent/CN108983769B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103900583A (en) * | 2012-12-25 | 2014-07-02 | 联想(北京)有限公司 | Device and method used for real-time positioning and map building |
| CN103886107A (en) * | 2014-04-14 | 2014-06-25 | 苏州市华天雄信息科技有限公司 | Robot locating and map building system based on ceiling image information |
| CN105825520A (en) * | 2015-01-08 | 2016-08-03 | 北京雷动云合智能技术有限公司 | Monocular SLAM (Simultaneous Localization and Mapping) method capable of creating large-scale map |
| CN104574421A (en) * | 2015-01-29 | 2015-04-29 | 北方工业大学 | Large-breadth small-overlapping-area high-precision multispectral image registration method and device |
| CN106355197A (en) * | 2016-08-24 | 2017-01-25 | 广东宝乐机器人股份有限公司 | Navigation image matching filtering method based on K-means clustering algorithm |
| WO2018089703A1 (en) * | 2016-11-09 | 2018-05-17 | The Texas A&M University System | Method and system for accurate long term simultaneous localization and mapping with absolute orientation sensing |
Non-Patent Citations (1)
| Title |
|---|
| 刘圣龙: "室内机器人的同步定位与建图方法的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109887003A (en) * | 2019-01-23 | 2019-06-14 | 亮风台（上海）信息科技有限公司 | A kind of method and apparatus initialized for carrying out three-dimensional tracking |
| CN109965797A (en) * | 2019-03-07 | 2019-07-05 | 深圳市愚公科技有限公司 | Generation method, sweeping robot control method and the terminal of sweeping robot map |
| CN109965797B (en) * | 2019-03-07 | 2021-08-24 | 深圳市愚公科技有限公司 | Floor sweeping robot map generation method, floor sweeping robot control method and terminal |
| CN112767818A (en) * | 2019-11-01 | 2021-05-07 | 北京初速度科技有限公司 | Map construction method and device |
| CN112767818B (en) * | 2019-11-01 | 2022-09-27 | 北京初速度科技有限公司 | Map construction method and device |
| CN113220818A (en) * | 2021-05-27 | 2021-08-06 | 南昌智能新能源汽车研究院 | Automatic mapping and high-precision positioning method for parking lot |
| CN113220818B (en) * | 2021-05-27 | 2023-04-07 | 南昌智能新能源汽车研究院 | Automatic mapping and high-precision positioning method for parking lot |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108983769B (en) | 2022-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108983769A (en) | Immediately the optimization method and device of positioning and map structuring | |
| CN111561923B (en) | SLAM (simultaneous localization and mapping) mapping method and system based on multi-sensor fusion | |
| US10706582B2 (en) | Real-time monocular structure from motion | |
| CN108090958B (en) | Robot synchronous positioning and map building method and system | |
| US11270148B2 (en) | Visual SLAM method and apparatus based on point and line features | |
| EP3977346A1 (en) | Simultaneous localization and mapping method, device, system and storage medium | |
| US9148650B2 (en) | Real-time monocular visual odometry | |
| CN102509074B (en) | Target identification method and device | |
| CN103136525B (en) | A kind of special-shaped Extended target high-precision locating method utilizing Generalized Hough Transform | |
| CN108369741A (en) | Method and system for registration data | |
| CN103093459B (en) | Utilize the method that airborne LiDAR point cloud data assisted image mates | |
| CN104197933B (en) | High magnitude slides enhancing and the extracting method of fixed star in a kind of range of telescope | |
| CN103700113A (en) | Method for detecting dim small moving target under downward-looking complicated background | |
| CN103644904A (en) | Visual navigation method based on SIFT (scale invariant feature transform) algorithm | |
| CN112950696A (en) | Navigation map generation method and generation device and electronic equipment | |
| CN108257155A (en) | A kind of extension target tenacious tracking point extracting method based on part and Global-Coupling | |
| Huang et al. | Joint ego-motion estimation using a laser scanner and a monocular camera through relative orientation estimation and 1-DoF ICP | |
| CN105466399A (en) | Quick semi-global dense matching method and device | |
| CN109358315B (en) | Auxiliary target indirect positioning method and system | |
| CN105574875B (en) | A kind of fish eye images dense stereo matching process based on polar geometry | |
| CN110335282A (en) | A kind of profile line segment feature extraction algorithm based on grid | |
| Majdik et al. | Micro air vehicle localization and position tracking from textured 3d cadastral models | |
| CN110264510A (en) | A method of image zooming-out depth of view information is acquired based on binocular | |
| CN103900563A (en) | Fast-moving star tracking matching identification method of star sensor | |
| Ou et al. | Targetless Lidar-camera Calibration via Cross-modality Structure Consistency |
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 |