CN104424637B - The method and smart machine of information processing - Google Patents
The method and smart machine of information processing Download PDFInfo
- Publication number
- CN104424637B CN104424637B CN201310392816.9A CN201310392816A CN104424637B CN 104424637 B CN104424637 B CN 104424637B CN 201310392816 A CN201310392816 A CN 201310392816A CN 104424637 B CN104424637 B CN 104424637B
- Authority
- CN
- China
- Prior art keywords
- information
- smart machine
- sub
- positional
- positional information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
Abstract
The invention discloses a kind of method of information processing, including:Obtain first information set and the second information aggregate;Corresponding smart machine first position information in the environment during each first information in calculating acquisition first information set;The first sub-information is obtained respectively from least two first information in first information set;The first object set is obtained respectively from the first sub-information at least two first information;The second place information of each the first object in the first object set in the first sub-information at least two first information is calculated respectively;The 3rd positional information of each the first object in the first object set in the first sub-information at least two first information is calculated respectively;The 3rd positional information to the object of identical first in the first object set in the first sub-information at least two first information for being calculated optimizes treatment.The invention also discloses a kind of smart machine.The embodiment of the present invention improves the efficiency and accuracy rate of characteristic point position acquisition of information.
Description
Technical field
The present invention relates to the information processing technology, more particularly to a kind of three-dimensional map information processing method based on smart machine
And smart machine.
Background technology
At present, the application with Based Intelligent Control in the industry is more extensive, and smart machine is just being applied to various technique productions
In.Carry out environmental information to create using smart machine is also one of commercial Application of smart machine.It is main in current environmental information
To be demarcated by determining the positional information of some characteristic points in smart machine collection image, pass through many spies for being extracted
Levy positional information a little and complete the collection of environmental information.
Because there is displacement error in smart machine, so it is determined that characteristic point positional information when, naturally also
There is error, and error once accumulating, the environmental information for being formed will be quite inaccurate.Therefore, environmental information wound is being carried out
When building, the positional information amendment to characteristic point will become quite important, once the positional information of the characteristic point for being gathered is inaccurate,
The practicality of the environmental information for being created will be deteriorated.
The content of the invention
In view of this, the main purpose of the embodiment of the present invention is the method and smart machine for providing a kind of information processing,
Can in the information of collection to the positional information of characteristic point in collection information be modified, obtain more accurately information.
To reach above-mentioned purpose, what the technical scheme of the embodiment of the present invention was realized in:
A kind of method of information processing, is applied in smart machine;Methods described includes:
Obtain first information set and the second information aggregate corresponding to the first information set;
Calculate the corresponding smart machine in the environment first when obtaining each first information in first information set
Confidence ceases;
The first sub-information is obtained respectively from least two first information in the first information set;
The first object set is obtained respectively from the first sub-information at least two first information;
According to the second information in second information aggregate corresponding with the first information in first information set, difference
Calculate in the first object set in the first sub-information at least two first information each first
The second place information of object;
According to the second place information and the smart machine second place information in the environment are calculated respectively
Of each the first object in the first object set in the first sub-information at least two first information
Three positional informations;
To described first pair in the first sub-information in described at least two first information for being calculated
As the 3rd positional information of the object of identical first in set optimizes treatment, the 4th of the object of the identical first is obtained
Positional information, as the positional information of the object of the identical first.
A kind of smart machine;Including:First acquisition unit, the first computing unit, second acquisition unit, the 3rd obtain single
Unit, the second computing unit, the 3rd computing unit and optimization processing unit, wherein:
First acquisition unit, for obtaining first information set and the second information collection corresponding to the first information set
Close;
First computing unit, for calculate obtain first information set in each first information when the corresponding smart machine
First position information in the environment;
Second acquisition unit, for being obtained respectively from least two first information in the first information set
Take the first sub-information;
3rd acquiring unit, for distinguishing from the first sub-information at least two first information
Obtain the first object set;
Second computing unit, for basis second information aggregate corresponding with the first information in first information set
In the second information, the first object in the first sub-information at least two first information is calculated respectively
The second place information of each the first object in set;
3rd computing unit, according to the second place information and the smart machine second place information in the environment
Each in the first object set in the first sub-information at least two first information is calculated respectively
3rd positional information of the first object;
Optimization processing unit, for the first son letter in the described at least two first information to being calculated
The 3rd positional information of the object of identical first optimizes treatment in first object set in breath, obtains described identical
The first object the 4th positional information, as the positional information of the object of the identical first.
In embodiments of the invention, first information set and the second information collection corresponding to the first information set are obtained
Close;Corresponding smart machine first position letter in the environment during each first information in calculating acquisition first information set
Breath;The first sub-information is obtained respectively from least two first information in the first information set;From it is described to
The first object set is obtained respectively in the first sub-information in few more than two first information;According to first information collection
The second information in corresponding second information aggregate of the first information in conjunction, calculates at least two institute respectively
State the second place information of each the first object in the first object set in the first sub-information in the first information;According to institute
State second place information and the smart machine second place information in the environment and calculate described at least two respectively
3rd positional information of each the first object in the first object set in the first sub-information in the first information;To institute
It is identical in first object set in the first sub-information in the described at least two first information for calculating
The 3rd positional information of the first object optimize treatment, obtain the 4th positional information of the object of the identical first, make
It is the positional information of the object of the identical first.In the embodiment of the present invention, by continuously acquiring multiple first information, and extract
The first sub-information for coinciding in these multiple first information, obtains the first object respectively from the first sub-information for coinciding
Set, and calculate the positional information of all first objects in each first object set, then the positional information of the first object is entered
Row amendment, using the positional information of revised first object as first object positional information.The first object is obtained again
During positional information, carried out using parallel mode, so as to improve treatment effeciency;Because the present invention is obtained from multiple first information
The positional information of the object of multiple identicals first, and positional information to these the first objects is modified, first for being calculated
The positional information of object is more accurate, when the positional information based on the first object creates environmental information, the more practical valency of environmental information
Value.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the method for the information processing of the embodiment of the present invention one;
Fig. 2 is the schematic flow sheet of the method for the information processing of the embodiment of the present invention two;
Fig. 3 is the schematic flow sheet of the method for the information processing of the embodiment of the present invention three;
Fig. 4 is the schematic flow sheet of the method for the information processing of the embodiment of the present invention four;
Fig. 5 is the composition structural representation of the smart machine of the embodiment of the present invention;
Fig. 6 is the composition structural representation of the 3rd computing unit in the smart machine of the embodiment of the present invention;
Fig. 7 is the composition structural representation of optimization processing unit in the smart machine of the embodiment of the present invention;
Fig. 8 is the composition structural representation of the first computing unit in the smart machine of the embodiment of the present invention.
Specific embodiment
The embodiment of the present invention, it is necessary to smart machine during primarily directed to current context information such as three-dimensional map information creating
As robot is constantly calculated and corrected to the positional information of acquired characteristic point, but to the positional information calculation of characteristic point
And the mode of correction is relatively more, the mode for being calculated is more numerous and diverse, and many calculations need to be aided with many special equipments, cause
It is relatively costly.
The embodiment of the present invention is proposed primarily directed to above-mentioned technical problem, by continuously acquiring multiple first information, and
The first sub-information for coinciding in these multiple first information is extracted, first is obtained respectively from the first sub-information for coinciding
Object set, and the positional information of all first objects in each first object set is calculated, then to the position letter of the first object
Breath is modified, using the positional information of revised first object as first object positional information.
It is by the following examples and referring to the drawings, right to make the object, technical solutions and advantages of the present invention become more apparent
The present invention is further described.
Fig. 1 is the schematic flow sheet of the method for the information processing of the embodiment of the present invention one, as shown in figure 1, in the present invention one
In individual preferred embodiment, the method for information processing is comprised the following steps:
The method of the information processing of the present embodiment is applied in smart machine, and such as smart machine can be intelligence machine
People, the intelligent robot has image acquisition units;Used as a preferred exemplary, the image acquisition units of the embodiment of the present invention can
Think camera, used as a kind of example, image acquisition units can also be special IMAQ and analysis system, such as can be
Image sensor, RGB image sensor etc..The smart machine also has image-capable.
Step 101, obtains first information set and the second information aggregate corresponding to the first information set;
In this example, the first information set includes:The set of the image information frame of the smart machine collection;
Second information aggregate includes:Corresponding to the set of the depth information of each image information frame.
Specifically, the image acquisition units by being set on the smart machine gather continuous image information frame, these companies
Continuous image information frame constitutes image information frame set, as first information set.
Corresponding to each image information frame, also to the depth information that there should be the image information frame, depth information set is formed.
Step 102, calculate obtain first information set in each first information when the corresponding smart machine in the environment
First position information;
The first position information includes:Positional information of the smart machine in global coordinate system;
In this step, when being acquired to image information frame, the position of smart machine when also to collection image information frame
Confidence breath is calculated.
Step 103, obtains the first son respectively from least two first information in the first information set
Information;
In this example, by continuously acquiring multiple first information, and coinciding in these multiple first information is extracted
First sub-information, obtains the first object set respectively from the first sub-information for coinciding.Specifically, by the figure of smart machine
As collecting unit obtains continuous image information frame, the phase of these image information frames is obtained from multiple continuous image information frames
Mutual intersection.These continuous multiple images information can be determined by the positional information of foregoing each image information frame
The intersection of frame, to obtain identical characteristic point from the image information frame for overlapping.
Step 104, first is obtained from the first sub-information at least two first information respectively
Object set;
First object set includes:All characteristic points in image information frame;
In this example, the first object can be smart machine by the spy of the object in image acquisition units institute viewfinder image
Levy a little;In this example, characteristic point can be the imaging point at any position on any object.As object edge point,
Can be the point inconsistent with the characteristic of the object on object, such as black splotch on white object, raised point on object,
Recessed point on object, the rust spot on metal object, the peel point on object table finish paint body etc..
Characteristic point also has many attribute informations such as color(RGB)Information, positional information, feature point description information etc..Feature
Description information is the information of the attribute of characteristic feature point, and characteristic point can be uniquely determined by characterization information.Specifically, may be used
Information and the spy of characteristic point of each characteristic point in set of characteristic points are determined by SIFT algorithms, SURF algorithm or ORB algorithms etc.
Levy description information;One characteristic point of characterization information energy unique mark of characteristic point, it is the data of 64 dimensions or 128 dimensions, when
So, further divided with to characteristic point attribute, the characterization information of characteristic feature point can be the parameter of other dimensions.
As an example, characterization information can include yardstick and rotation.
Characteristic point is obtained from the image information of the intersection of continuous multiple images information frame, due to the figure being based on
As information is identical, therefore, should be identical from the characteristic point accessed by the intersection in each image information frame,
Using acquired characteristic point as set of characteristic points.
Step 105, according to second in second information aggregate corresponding with the first information in first information set
Information, in calculating the first object set in the first sub-information at least two first information respectively
The second place information of each the first object;
Wherein, the second place information includes:Relative position information of first object relative to the smart machine;
After the characteristic point of the image information of intersection of continuous multiple images information frame is determined, believed according to image
Cease frame depth information, calculate respectively the intersection of continuous multiple images information frame each characteristic point relative to intelligence set
The positional information of standby image acquisition units.
Step 106, according to the second place information and the smart machine second place information difference in the environment
Calculate in the first object set in the first sub-information at least two first information each first
3rd positional information of object;
Wherein, the 3rd positional information includes:Positional information of the characteristic point in global coordinate system;
The image relative to smart machine for calculating each characteristic point of the intersection of continuous multiple images information frame is adopted
After collecting the positional information of unit, according to the position of the moment corresponding smart machine in global coordinate system of collection image information frame
Information, calculates the positional information of each characteristic point in global coordinate system of the intersection of continuous multiple images information frame.
Step 107, to the institute in the first sub-information in described at least two first information for being calculated
The 3rd positional information for stating the object of identical first in the first object set optimizes treatment, obtains the identical first pair
4th positional information of elephant, as the positional information of the object of the identical first.
Because smart machine understands existence position information calculation error in moving process, therefore, by true before this step
Can there is larger error, the 3rd position of the characteristic point that this example need to be calculated abovementioned steps in the positional information of fixed characteristic point
Information optimizes treatment, specifically, to each position information of calculated same characteristic features point, asks arithmetic average or weighted average
Deng as the final position information of the same characteristic features point.
In this example, the judgement of same characteristic features point can be carried out by the RGB information of characteristic point or characterization information etc.
It is determined that.For example when the corresponding RGB information of characteristic point is identical, determine that characteristic point is identical, or, the feature description letter of characteristic point
When each single item in breath is matched, determine that characteristic point is identical.
Fig. 2 is the schematic flow sheet of the method for the information processing of the embodiment of the present invention two, as shown in Fig. 2 in the present invention one
In individual preferred embodiment, the method for information processing is comprised the following steps:
The method of the information processing of the present embodiment is applied in smart machine, and such as smart machine can be intelligence machine
People, the intelligent robot has image acquisition units;Used as a preferred exemplary, the image acquisition units of the embodiment of the present invention can
Think camera, used as a kind of example, image acquisition units can also be special IMAQ and analysis system, such as can be
Image sensor, RGB image sensor etc..The smart machine also has image-capable.
Step 201, obtains first information set and the second information aggregate corresponding to the first information set;
In this example, the first information set includes:The set of the image information frame of the smart machine collection;
Second information aggregate includes:Corresponding to the set of the depth information of each image information frame.
Specifically, the image acquisition units by being set on the smart machine gather continuous image information frame, these companies
Continuous image information frame constitutes image information frame set, as first information set.
Corresponding to each image information frame, also to the depth information that there should be the image information frame, depth information set is formed.
Step 202, calculate obtain first information set in each first information when the corresponding smart machine in the environment
First position information;
The first position information includes:Positional information of the smart machine in global coordinate system;
In this step, when being acquired to image information frame, the position of smart machine when also to collection image information frame
Confidence breath is calculated.
Step 203, obtains the first son respectively from least two first information in the first information set
Information;
In this example, by continuously acquiring multiple first information, and coinciding in these multiple first information is extracted
First sub-information, obtains the first object set respectively from the first sub-information for coinciding.Specifically, by the figure of smart machine
As collecting unit obtains continuous image information frame, the phase of these image information frames is obtained from multiple continuous image information frames
Mutual intersection.These continuous multiple images information can be determined by the positional information of foregoing each image information frame
The intersection of frame, to obtain identical characteristic point from the image information frame for overlapping.
Step 204, first is obtained from the first sub-information at least two first information respectively
Object set;
First object set includes:All characteristic points in image information frame;
In this example, the first object can be smart machine by the spy of the object in image acquisition units institute viewfinder image
Levy a little;In this example, characteristic point can be the imaging point at any position on any object.As object edge point,
Can be the point inconsistent with the characteristic of the object on object, such as black splotch on white object, raised point on object,
Recessed point on object, the rust spot on metal object, the peel point on object table finish paint body etc..
Characteristic point also has many attribute informations such as color(RGB)Information, positional information, feature point description information etc..Feature
Description information is the information of the attribute of characteristic feature point, and characteristic point can be uniquely determined by characterization information.Specifically, may be used
Information and the spy of characteristic point of each characteristic point in set of characteristic points are determined by SIFT algorithms, SURF algorithm or ORB algorithms etc.
Levy description information;One characteristic point of characterization information energy unique mark of characteristic point, it is the data of 64 dimensions or 128 dimensions, when
So, further divided with to characteristic point attribute, the characterization information of characteristic feature point can be the parameter of other dimensions.
As an example, characterization information can include yardstick and rotation.
Characteristic point is obtained from the image information of the intersection of continuous multiple images information frame, due to the figure being based on
As information is identical, therefore, should be identical from the characteristic point accessed by the intersection in each image information frame,
Using acquired characteristic point as set of characteristic points.
Step 205, according to second in second information aggregate corresponding with the first information in first information set
Information, in calculating the first object set in the first sub-information at least two first information respectively
The second place information of each the first object;
Wherein, the second place information includes:Relative position information of first object relative to the smart machine;
After the characteristic point of the image information of intersection of continuous multiple images information frame is determined, believed according to image
Cease frame depth information, calculate respectively the intersection of continuous multiple images information frame each characteristic point relative to intelligence set
The positional information of standby image acquisition units.
Step 206, according to the second place information and the smart machine second place information difference in the environment
Calculate in the first object set in the first sub-information at least two first information each first
3rd positional information of object;
Wherein, the 3rd positional information includes:Positional information of the characteristic point in global coordinate system;
Specifically, the first object set in the first sub-information at least two first information is calculated
3rd positional information of each the first object in conjunction, including:
By first object set in the first sub-information in the first information in each described first information set
In all first sets spatially position is divided at least two part;
In the first object set in the first sub-information at least two first information
The second place information and the smart machine of each the first object first position information in the environment, described in parallel computation
Of each the first object in the first object set in the first sub-information at least two first information
Three positional informations.
The image relative to smart machine for calculating each characteristic point of the intersection of continuous multiple images information frame is adopted
After collecting the positional information of unit, according to the position of the moment corresponding smart machine in global coordinate system of collection image information frame
Information, calculates the positional information of each characteristic point in global coordinate system of the intersection of continuous multiple images information frame.
For the set of characteristic points extracted from each image information frame, multiple areas are divided into by the locus of characteristic point
Domain, is such as divided into four regions, eight regions, 16 regions.It is more uniform to ensure the ratio that set of characteristic points is divided, can press
The coordinate value of all characteristic points in set of characteristic points, determines that the square that can cover all characteristic points in set of characteristic points is empty
Between, square space is divided into four sub- squares of identical;Each can be concurrently calculated in units of every sub- square
The positional information of the characteristic point in sub- square.
It should be noted that to ensure that each characteristic point in set of characteristic points is calculated with being all not missed, being divided
Sub- square between can have the region of coincidence.
It should be noted that this example be set of characteristic points carried out by taking square as an example region division, or
Other dividing modes, are such as divided by the quantity of characteristic point in set of characteristic points.In view of the establishment feature of environmental information,
Region division preferably is carried out to set of characteristic points by square space.
Step 207, to the institute in the first sub-information in described at least two first information for being calculated
The 3rd positional information for stating the object of identical first in the first object set optimizes treatment, obtains the identical first pair
4th positional information of elephant, as the positional information of the object of the identical first.
Because smart machine understands existence position information calculation error in moving process, therefore, by true before this step
Can there is larger error, the 3rd position of the characteristic point that this example need to be calculated abovementioned steps in the positional information of fixed characteristic point
Information optimizes treatment, specifically, to each position information of calculated same characteristic features point, asks arithmetic average or weighted average
Deng as the final position information of the same characteristic features point.
In this example, the judgement of same characteristic features point can be carried out by the RGB information of characteristic point or characterization information etc.
It is determined that.For example when the corresponding RGB information of characteristic point is identical, determine that characteristic point is identical, or, the feature description letter of characteristic point
When each single item in breath is matched, determine that characteristic point is identical.
Fig. 3 is the schematic flow sheet of the method for the information processing of the embodiment of the present invention three, as shown in figure 3, in the present invention one
In individual preferred embodiment, the method for information processing is comprised the following steps:
The method of the information processing of the present embodiment is applied in smart machine, and such as smart machine can be intelligence machine
People, the intelligent robot has image acquisition units;Used as a preferred exemplary, the image acquisition units of the embodiment of the present invention can
Think camera, used as a kind of example, image acquisition units can also be special IMAQ and analysis system, such as can be
Image sensor, RGB image sensor etc..The smart machine also has image-capable.
Step 301, obtains first information set and the second information aggregate corresponding to the first information set;
In this example, the first information set includes:The set of the image information frame of the smart machine collection;
Second information aggregate includes:Corresponding to the set of the depth information of each image information frame.
Specifically, the image acquisition units by being set on the smart machine gather continuous image information frame, these companies
Continuous image information frame constitutes image information frame set, as first information set.
Corresponding to each image information frame, also to the depth information that there should be the image information frame, depth information set is formed.
Step 302, calculate obtain first information set in each first information when the corresponding smart machine in the environment
First position information;
The first position information includes:Positional information of the smart machine in global coordinate system;
In this step, when being acquired to image information frame, the position of smart machine when also to collection image information frame
Confidence breath is calculated.
Step 303, obtains the first son respectively from least two first information in the first information set
Information;
In this example, by continuously acquiring multiple first information, and coinciding in these multiple first information is extracted
First sub-information, obtains the first object set respectively from the first sub-information for coinciding.Specifically, by the figure of smart machine
As collecting unit obtains continuous image information frame, the phase of these image information frames is obtained from multiple continuous image information frames
Mutual intersection.These continuous multiple images information can be determined by the positional information of foregoing each image information frame
The intersection of frame, to obtain identical characteristic point from the image information frame for overlapping.
Step 304, first is obtained from the first sub-information at least two first information respectively
Object set;
First object set includes:All characteristic points in image information frame;
In this example, the first object can be smart machine by the spy of the object in image acquisition units institute viewfinder image
Levy a little;In this example, characteristic point can be the imaging point at any position on any object.As object edge point,
Can be the point inconsistent with the characteristic of the object on object, such as black splotch on white object, raised point on object,
Recessed point on object, the rust spot on metal object, the peel point on object table finish paint body etc..
Characteristic point also has many attribute informations such as color(RGB)Information, positional information, feature point description information etc..Feature
Description information is the information of the attribute of characteristic feature point, and characteristic point can be uniquely determined by characterization information.Specifically, may be used
Information and the spy of characteristic point of each characteristic point in set of characteristic points are determined by SIFT algorithms, SURF algorithm or ORB algorithms etc.
Levy description information;One characteristic point of characterization information energy unique mark of characteristic point, it is the data of 64 dimensions or 128 dimensions, when
So, further divided with to characteristic point attribute, the characterization information of characteristic feature point can be the parameter of other dimensions.
As an example, characterization information can include yardstick and rotation.
Characteristic point is obtained from the image information of the intersection of continuous multiple images information frame, due to the figure being based on
As information is identical, therefore, should be identical from the characteristic point accessed by the intersection in each image information frame,
Using acquired characteristic point as set of characteristic points.
Step 305, according to second in second information aggregate corresponding with the first information in first information set
Information, in calculating the first object set in the first sub-information at least two first information respectively
The second place information of each the first object;
Wherein, the second place information includes:Relative position information of first object relative to the smart machine;
After the characteristic point of the image information of intersection of continuous multiple images information frame is determined, believed according to image
Cease frame depth information, calculate respectively the intersection of continuous multiple images information frame each characteristic point relative to intelligence set
The positional information of standby image acquisition units.
Step 306, according to the second place information and the smart machine second place information difference in the environment
Calculate in the first object set in the first sub-information at least two first information each first
3rd positional information of object;
Wherein, the 3rd positional information includes:Positional information of the characteristic point in global coordinate system;
Specifically, the first object set in the first sub-information at least two first information is calculated
3rd positional information of each the first object in conjunction, including:
By first object set in the first sub-information in the first information in each described first information set
In all first sets spatially position is divided at least two part;
In the first object set in the first sub-information at least two first information
The second place information and the smart machine of each the first object first position information in the environment, described in parallel computation
Of each the first object in the first object set in the first sub-information at least two first information
Three positional informations.
The image relative to smart machine for calculating each characteristic point of the intersection of continuous multiple images information frame is adopted
After collecting the positional information of unit, according to the position of the moment corresponding smart machine in global coordinate system of collection image information frame
Information, calculates the positional information of each characteristic point in global coordinate system of the intersection of continuous multiple images information frame.
For the set of characteristic points extracted from each image information frame, multiple areas are divided into by the locus of characteristic point
Domain, is such as divided into four regions, eight regions, 16 regions.It is more uniform to ensure the ratio that set of characteristic points is divided, can press
The coordinate value of all characteristic points in set of characteristic points, determines that the square that can cover all characteristic points in set of characteristic points is empty
Between, square space is divided into four sub- squares of identical;Each can be concurrently calculated in units of every sub- square
The positional information of the characteristic point in sub- square.
It should be noted that to ensure that each characteristic point in set of characteristic points is calculated with being all not missed, being divided
Sub- square between can have the region of coincidence.
It should be noted that this example be set of characteristic points carried out by taking square as an example region division, or
Other dividing modes, are such as divided by the quantity of characteristic point in set of characteristic points.In view of the establishment feature of environmental information,
Region division preferably is carried out to set of characteristic points by square space.
Step 307, to the institute in the first sub-information in described at least two first information for being calculated
The 3rd positional information for stating the object of identical first in the first object set optimizes treatment, obtains the identical first pair
4th positional information of elephant, as the positional information of the object of the identical first.
Specifically, optimizing treatment to the 3rd positional information includes:
It is the identical first in the first sub-information in described at least two first information for being calculated
3rd positional information of object sets up Gauss model;
Determine the Gaussian distribution feature of the 3rd positional information of the object of the identical first;
The Gaussian distribution feature of the 3rd positional information according to the object of the identical first screens out variance beyond setting threshold
The 3rd positional information beyond the object of the identical first of value;
Calculate the average value of the 3rd positional information of the remaining object of the identical first.
Here, average value includes weighted average or arithmetic mean of instantaneous value.
Because smart machine understands existence position information calculation error in moving process, therefore, by true before this step
Can there is larger error, the 3rd position of the characteristic point that this example need to be calculated abovementioned steps in the positional information of fixed characteristic point
Information optimizes treatment, specifically, to each position information of calculated same characteristic features point, asks arithmetic average or weighted average
Deng as the final position information of the same characteristic features point.
Gauss model is set up by the world coordinates positional information to same characteristic features point, the overall situation of same characteristic features point is determined
Co-ordinate position information Gaussian Profile state, and screen out the world coordinates positional information of characteristic point away from the equal line of Gaussian Profile.
These can be regarded as the larger result of calculation of error away from the world coordinates positional information of the characteristic point of the equal line of Gaussian Profile, such as may be used
Can there is mistake in computation etc..
In this example, the judgement of same characteristic features point can be carried out by the RGB information of characteristic point or characterization information etc.
It is determined that.For example when the corresponding RGB information of characteristic point is identical, determine that characteristic point is identical, or, the feature description letter of characteristic point
When each single item in breath is matched, determine that characteristic point is identical.
Fig. 4 is the schematic flow sheet of the method for the information processing of the embodiment of the present invention four, as shown in figure 4, in the present invention one
In individual preferred embodiment, the method for information processing is comprised the following steps:
The method of the information processing of the present embodiment is applied in smart machine, and such as smart machine can be intelligence machine
People, the intelligent robot has image acquisition units;Used as a preferred exemplary, the image acquisition units of the embodiment of the present invention can
Think camera, used as a kind of example, image acquisition units can also be special IMAQ and analysis system, such as can be
Image sensor, RGB image sensor etc..The smart machine also has image-capable.
Step 401, obtains first information set and the second information aggregate corresponding to the first information set;
In this example, the first information set includes:The set of the image information frame of the smart machine collection;
Second information aggregate includes:Corresponding to the set of the depth information of each image information frame.
Specifically, the image acquisition units by being set on the smart machine gather continuous image information frame, these companies
Continuous image information frame constitutes image information frame set, as first information set.
Corresponding to each image information frame, also to the depth information that there should be the image information frame, depth information set is formed.
Step 402, calculate obtain first information set in each first information when the corresponding smart machine in the environment
First position information;
The first position information includes:Positional information of the smart machine in global coordinate system;
In this step, when being acquired to image information frame, the position of smart machine when also to collection image information frame
Confidence breath is calculated.
Specifically, it is determined that second place information, including:
Respectively to the smart machine move horizontally direction and vertically move direction enter row distance correction;
It is determined that obtaining the time information of all first information in first information set;
According to the first position information before smart machine distance correction, determine time information, move horizontally direction
And vertically move the correction distance in direction, calculate the smart machine in obtaining first information set all first information when
The first position information at quarter.
Specifically, for smart machine, the kinetic control system for determining smart machine displacement, example can be typically provided with
Incremental rotary encoder, the real time measure intelligent robot such as can be respectively mounted in the left and right sidesing driving wheel both sides of intelligent robot
The distance of left and right wheels walking in unit interval, and can show that the pose of robot is sat by calculating by the travel distance of two-wheeled
Mark.And the angle value in intelligent robot pose is calculated, by the principle of Differential Driving intelligent robot, you can determine intelligence
Can positional information of the robot in its local environment.
Step 403, obtains the first son respectively from least two first information in the first information set
Information;
In this example, by continuously acquiring multiple first information, and coinciding in these multiple first information is extracted
First sub-information, obtains the first object set respectively from the first sub-information for coinciding.Specifically, by the figure of smart machine
As collecting unit obtains continuous image information frame, the phase of these image information frames is obtained from multiple continuous image information frames
Mutual intersection.These continuous multiple images information can be determined by the positional information of foregoing each image information frame
The intersection of frame, to obtain identical characteristic point from the image information frame for overlapping.
Step 404, first is obtained from the first sub-information at least two first information respectively
Object set;
First object set includes:All characteristic points in image information frame;
In this example, the first object can be smart machine by the spy of the object in image acquisition units institute viewfinder image
Levy a little;In this example, characteristic point can be the imaging point at any position on any object.As object edge point,
Can be the point inconsistent with the characteristic of the object on object, such as black splotch on white object, raised point on object,
Recessed point on object, the rust spot on metal object, the peel point on object table finish paint body etc..
Characteristic point also has many attribute informations such as color(RGB)Information, positional information, feature point description information etc..Feature
Description information is the information of the attribute of characteristic feature point, and characteristic point can be uniquely determined by characterization information.Specifically, may be used
Information and the spy of characteristic point of each characteristic point in set of characteristic points are determined by SIFT algorithms, SURF algorithm or ORB algorithms etc.
Levy description information;One characteristic point of characterization information energy unique mark of characteristic point, it is the data of 64 dimensions or 128 dimensions, when
So, further divided with to characteristic point attribute, the characterization information of characteristic feature point can be the parameter of other dimensions.
As an example, characterization information can include yardstick and rotation.
Characteristic point is obtained from the image information of the intersection of continuous multiple images information frame, due to the figure being based on
As information is identical, therefore, should be identical from the characteristic point accessed by the intersection in each image information frame,
Using acquired characteristic point as set of characteristic points.
Step 405, according to second in second information aggregate corresponding with the first information in first information set
Information, in calculating the first object set in the first sub-information at least two first information respectively
The second place information of each the first object;
Wherein, the second place information includes:Relative position information of first object relative to the smart machine;
After the characteristic point of the image information of intersection of continuous multiple images information frame is determined, believed according to image
Cease frame depth information, calculate respectively the intersection of continuous multiple images information frame each characteristic point relative to intelligence set
The positional information of standby image acquisition units.
Step 406, according to the second place information and the smart machine second place information difference in the environment
Calculate in the first object set in the first sub-information at least two first information each first
3rd positional information of object;
Wherein, the 3rd positional information includes:Positional information of the characteristic point in global coordinate system;
Specifically, the first object set in the first sub-information at least two first information is calculated
3rd positional information of each the first object in conjunction, including:
By first object set in the first sub-information in the first information in each described first information set
In all first sets spatially position is divided at least two part;
In the first object set in the first sub-information at least two first information
The second place information and the smart machine of each the first object first position information in the environment, described in parallel computation
Of each the first object in the first object set in the first sub-information at least two first information
Three positional informations.
The image relative to smart machine for calculating each characteristic point of the intersection of continuous multiple images information frame is adopted
After collecting the positional information of unit, according to the position of the moment corresponding smart machine in global coordinate system of collection image information frame
Information, calculates the positional information of each characteristic point in global coordinate system of the intersection of continuous multiple images information frame.
For the set of characteristic points extracted from each image information frame, multiple areas are divided into by the locus of characteristic point
Domain, is such as divided into four regions, eight regions, 16 regions.It is more uniform to ensure the ratio that set of characteristic points is divided, can press
The coordinate value of all characteristic points in set of characteristic points, determines that the square that can cover all characteristic points in set of characteristic points is empty
Between, square space is divided into four sub- squares of identical;Each can be concurrently calculated in units of every sub- square
The positional information of the characteristic point in sub- square.
It should be noted that to ensure that each characteristic point in set of characteristic points is calculated with being all not missed, being divided
Sub- square between can have the region of coincidence.
This example is to have carried out region division, or other dividing modes to set of characteristic points by taking square as an example,
Such as divided by the quantity of characteristic point in set of characteristic points.In view of the establishment feature of environmental information, preferably by square
Space carries out region division to set of characteristic points.
Step 407, to the institute in the first sub-information in described at least two first information for being calculated
The 3rd positional information for stating the object of identical first in the first object set optimizes treatment, obtains the identical first pair
4th positional information of elephant, as the positional information of the object of the identical first.
Specifically, optimizing treatment to the 3rd positional information includes:
It is the identical first in the first sub-information in described at least two first information for being calculated
3rd positional information of object sets up Gauss model;
Determine the Gaussian distribution feature of the 3rd positional information of the object of the identical first;
The Gaussian distribution feature of the 3rd positional information according to the object of the identical first screens out variance beyond setting threshold
The 3rd positional information beyond the object of the identical first of value;
Calculate the average value of the 3rd positional information of the remaining object of the identical first.
Here, average value includes weighted average or arithmetic mean of instantaneous value.
Because smart machine understands existence position information calculation error in moving process, therefore, by true before this step
Can there is larger error, the 3rd position of the characteristic point that this example need to be calculated abovementioned steps in the positional information of fixed characteristic point
Information optimizes treatment, specifically, to each position information of calculated same characteristic features point, asks arithmetic average or weighted average
Deng as the final position information of the same characteristic features point.
Gauss model is set up by the world coordinates positional information to same characteristic features point, the overall situation of same characteristic features point is determined
Co-ordinate position information Gaussian Profile state, and screen out the world coordinates positional information of characteristic point away from the equal line of Gaussian Profile.
These can be regarded as the larger result of calculation of error away from the world coordinates positional information of the characteristic point of the equal line of Gaussian Profile, such as may be used
Can there is mistake in computation etc..
In this example, the judgement of same characteristic features point can be carried out by the RGB information of characteristic point or characterization information etc.
It is determined that.For example when the corresponding RGB information of characteristic point is identical, determine that characteristic point is identical, or, the feature description letter of characteristic point
When each single item in breath is matched, determine that characteristic point is identical.
Fig. 5 is the composition structural representation of the smart machine of the embodiment of the present invention, and in the embodiment of the present invention, smart machine can
Think intelligent robot, the intelligent robot has image acquisition units;As a preferred exemplary, the figure of the embodiment of the present invention
As collecting unit can be camera, used as a kind of example, image acquisition units can also be special IMAQ and analysis
System, such as can be image sensor, RGB image sensor.The smart machine also has image-capable.Such as Fig. 5 institutes
Show, the smart machine of the embodiment of the present invention includes:First acquisition unit 50, the first computing unit 51, second acquisition unit 52,
Three acquiring units 53, the second computing unit 54, the 3rd computing unit 55 and optimization processing unit 56, wherein:
First acquisition unit 50, for obtaining first information set and the second information corresponding to the first information set
Set;
First computing unit 51, for the corresponding intelligence to set during each first information in calculating acquisition first information set
Standby first position information in the environment;
Second acquisition unit 52, for distinguishing from least two first information in the first information set
Obtain the first sub-information;
3rd acquiring unit 53, for dividing from the first sub-information at least two first information
The first object set is not obtained;
Second computing unit 54, for basis the second information collection corresponding with the first information in first information set
The second information in conjunction, calculates the first couple in the first sub-information at least two first information respectively
As the second place information of the object of each in set first;
3rd computing unit 55, according to the second place information and the smart machine second confidence in the environment
It is every in the first object set in the first sub-information that breath calculates at least two first information respectively
3rd positional information of individual first object;
Optimization processing unit 56, for the first son in the described at least two first information to being calculated
The 3rd positional information of the object of identical first optimizes treatment in first object set in information, obtains the phase
4th positional information of the first same object, as the positional information of the object of the identical first.
In this example, the first information set includes:The set of the image information frame of the smart machine collection;
Second information aggregate includes:Corresponding to the set of the depth information of each image information frame;
The first position information includes:Positional information of the smart machine in global coordinate system;
The second place information includes:Relative position information of first object relative to the smart machine;
First object includes:The characteristic point extracted from described image information frame;
First object set includes:All characteristic points in image information frame;
3rd positional information includes:Positional information of the characteristic point in global coordinate system.
In this example, the first object can be smart machine by the spy of the object in image acquisition units institute viewfinder image
Levy a little;In this example, characteristic point can be the imaging point at any position on any object.As object edge point,
Can be the point inconsistent with the characteristic of the object on object, such as black splotch on white object, raised point on object,
Recessed point on object, the rust spot on metal object, the peel point on object table finish paint body etc..
Characteristic point also has many attribute informations such as color(RGB)Information, positional information, feature point description information etc..Feature
Description information is the information of the attribute of characteristic feature point, and characteristic point can be uniquely determined by characterization information.Specifically, may be used
Information and the spy of characteristic point of each characteristic point in set of characteristic points are determined by SIFT algorithms, SURF algorithm or ORB algorithms etc.
Levy description information;One characteristic point of characterization information energy unique mark of characteristic point, it is the data of 64 dimensions or 128 dimensions, when
So, further divided with to characteristic point attribute, the characterization information of characteristic feature point can be the parameter of other dimensions.
As an example, characterization information can include yardstick and rotation.
Used as a kind of preferred embodiment, the composition of the 3rd computing unit 55 of the embodiment of the present invention is as shown in Figure 6;Fig. 6 is
The composition structural representation of the 3rd computing unit in the smart machine of the embodiment of the present invention, as shown in fig. 6, the embodiment of the present invention
3rd computing unit 55 includes:Subelement 550 and parallel computation subelement 551 are divided, wherein:
Subelement 550 is divided, for by the first sub-information in the first information in each described first information set
First object set in all first sets spatially position is divided at least two part;
Parallel computation subelement 551, for the first son letter at least two first information
The second place information and the smart machine of each the first object in the first object set in breath in the environment
One positional information, the first object set in the first sub-information at least two first information described in parallel computation
3rd positional information of each the first object in conjunction.
The image relative to smart machine for calculating each characteristic point of the intersection of continuous multiple images information frame is adopted
After collecting the positional information of unit, according to the position of the moment corresponding smart machine in global coordinate system of collection image information frame
Information, calculates the positional information of each characteristic point in global coordinate system of the intersection of continuous multiple images information frame.
For the set of characteristic points extracted from each image information frame, multiple areas are divided into by the locus of characteristic point
Domain, is such as divided into four regions, eight regions, 16 regions.It is more uniform to ensure the ratio that set of characteristic points is divided, can press
The coordinate value of all characteristic points in set of characteristic points, determines that the square that can cover all characteristic points in set of characteristic points is empty
Between, square space is divided into four sub- squares of identical;Each can be concurrently calculated in units of every sub- square
The positional information of the characteristic point in sub- square.
It should be noted that to ensure that each characteristic point in set of characteristic points is calculated with being all not missed, being divided
Sub- square between can have the region of coincidence.
It should be noted that this example be set of characteristic points carried out by taking square as an example region division, or
Other dividing modes, are such as divided by the quantity of characteristic point in set of characteristic points.In view of the establishment feature of environmental information,
Region division preferably is carried out to set of characteristic points by square space.
Used as a kind of preferred embodiment, the composition of the optimization processing unit 56 of the embodiment of the present invention is as shown in Figure 7;Fig. 7 is
The composition structural representation of optimization processing unit 56 in the smart machine of the embodiment of the present invention, as shown in fig. 7, the embodiment of the present invention
Optimization processing unit include:Subelement 560, determination subelement 561, screening subelement 562 and computation subunit 563 are set up,
Wherein:
Subelement 560 is set up, for being the first son in described at least two first information for being calculated
3rd positional information of the object of identical first in information sets up Gauss model;
Determination subelement 561, the Gaussian distribution feature of the 3rd positional information for determining the object of the identical first;
Screening subelement 562, for the Gaussian distribution feature of the 3rd positional information according to the object of the identical first
Screen out threeth positional information beyond the identical first object of the variance beyond given threshold;
Computation subunit 563, the average value of the 3rd positional information for calculating the remaining object of the identical first.
Here, average value includes weighted average or arithmetic mean of instantaneous value.
Because smart machine understands existence position information calculation error in moving process, therefore, by the 3rd computing unit
Can there is larger error, the 3rd of the characteristic point that this example need to be calculated abovementioned steps the in the positional information of 55 characteristic points for calculating
Positional information optimizes treatment, specifically, to each position information of calculated same characteristic features point, asks arithmetic average or weighting flat
Equalization, as the final position information of the same characteristic features point.
Gauss model is set up by the world coordinates positional information to same characteristic features point, the overall situation of same characteristic features point is determined
Co-ordinate position information Gaussian Profile state, and screen out the world coordinates positional information of characteristic point away from the equal line of Gaussian Profile.
These can be regarded as the larger result of calculation of error away from the world coordinates positional information of the characteristic point of the equal line of Gaussian Profile, such as may be used
Can there is mistake in computation etc..
In this example, the judgement of same characteristic features point can be carried out by the RGB information of characteristic point or characterization information etc.
It is determined that.For example when the corresponding RGB information of characteristic point is identical, determine that characteristic point is identical, or, the feature description letter of characteristic point
When each single item in breath is matched, determine that characteristic point is identical.
Used as a kind of preferred embodiment, the composition of the first computing unit 51 of the embodiment of the present invention is as shown in Figure 8;Fig. 8 is
The composition structural representation of the first computing unit 51 in the smart machine of the embodiment of the present invention, as shown in figure 8, the embodiment of the present invention
The first computing unit 51 include:Correction subelement 510, determination subelement 511 and computation subunit 512, wherein:
Correction subelement 510, for moving horizontally direction and vertically moving direction and carry out to the smart machine respectively
Distance correction;
Determination subelement 512, the time information for determining all first information in acquisition first information set;
Computation subunit 513, for the first position information according to the smart machine before correction, determines the moment
Information, the correction distance for moving horizontally direction and vertically moving direction, calculate the smart machine and are obtaining first information set
In all first information moment first position information.
Specifically, for smart machine, the kinetic control system for determining smart machine displacement, example can be typically provided with
Incremental rotary encoder, the real time measure intelligent robot such as can be respectively mounted in the left and right sidesing driving wheel both sides of intelligent robot
The distance of left and right wheels walking in unit interval, and can show that the pose of robot is sat by calculating by the travel distance of two-wheeled
Mark.And the angle value in intelligent robot pose is calculated, by the principle of Differential Driving intelligent robot, you can determine intelligence
Can positional information of the robot in its local environment.
It will be appreciated by those skilled in the art that each processing unit and its subelement in the smart machine of the embodiment of the present invention
Function, can refer to the associated description of the method for foregoing information processing and understands, in the smart machine of the embodiment of the present invention everywhere
Reason unit and its subelement, can be realized, it is also possible to pass through by realizing the analog circuit of the function described in the embodiment of the present invention
Perform the operation of the software on smart machine of the function described in the embodiment of the present invention and realize.
Between technical scheme described in the embodiment of the present invention, in the case where not conflicting, can be in any combination.
In several embodiments provided by the present invention, it should be understood that disclosed method and smart machine, Ke Yitong
Other modes are crossed to realize.Apparatus embodiments described above are only schematical, for example, the division of the unit, only
Only a kind of division of logic function, can have other dividing mode, such as when actually realizing:Multiple units or component can be tied
Close, or be desirably integrated into another system, or some features can be ignored, or do not perform.In addition, shown or discussed each group
Into part coupling each other or direct-coupling or communication connection can be by some interfaces, equipment or unit it is indirect
Coupling is communicated to connect, and can be electrical, machinery or other forms.
The above-mentioned unit that is illustrated as separating component can be or may not be it is physically separate, it is aobvious as unit
The part for showing can be or may not be physical location, you can with positioned at a place, it is also possible to be distributed to multiple network lists
In unit;Part or all of unit therein can be according to the actual needs selected to realize the purpose of this embodiment scheme.
In addition, each functional unit in various embodiments of the present invention can be fully integrated into a processing unit, also may be used
Being each unit individually as a unit, it is also possible to which two or more units are integrated in a unit;It is above-mentioned
Integrated unit can both be realized in the form of hardware, it would however also be possible to employ hardware adds the form of SFU software functional unit to realize.
One of ordinary skill in the art will appreciate that:Realizing all or part of step of above method embodiment can pass through
Programmed instruction related hardware is completed, and foregoing program can be stored in a computer read/write memory medium, the program
Upon execution, the step of including above method embodiment is performed;And foregoing storage medium includes:It is movable storage device, read-only
Memory(ROM, Read-Only Memory), random access memory(RAM, Random Access Memory), magnetic disc or
Person's CD etc. is various can be with the medium of store program codes.
Or, if the above-mentioned integrated unit of the embodiment of the present invention is using realization in the form of software function module and as independently
Production marketing or when using, it is also possible to storage is in a computer read/write memory medium.Based on such understanding, this hair
The part that the technical scheme of bright embodiment substantially contributes to prior art in other words can in the form of software product body
Reveal and, the computer software product is stored in a storage medium, including some instructions are used to so that a computer sets
It is standby(Can be personal computer, server or network equipment etc.)Perform the whole of each embodiment methods described of the invention
Or part.And foregoing storage medium includes:Movable storage device, read-only storage(ROM, Read-Only Memory), with
Machine accesses memory(RAM, Random Access Memory), magnetic disc or CD etc. are various can be with Jie of store program codes
Matter.
The above, specific embodiment only of the invention, but protection scope of the present invention is not limited thereto, and it is any
Those familiar with the art the invention discloses technical scope in, change or replacement can be readily occurred in, should all contain
Cover within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (8)
1. a kind of method of information processing, is applied in smart machine;Characterized in that, methods described includes:
Obtain first information set and the second information aggregate corresponding to the first information set;
Corresponding smart machine first position letter in the environment during each first information in calculating acquisition first information set
Breath;
The first sub-information is obtained respectively from least two first information in the first information set;
The first object set is obtained respectively from the first sub-information at least two first information;
According to the second information in second information aggregate corresponding with the first information in first information set, calculate respectively
Each first object in the first object set in the first sub-information at least two first information
Second place information;
According to the second place information and the smart machine second place information in the environment are calculated respectively at least
The 3rd of each the first object in the first object set in the first sub-information in more than two first information
Confidence ceases;
To first object set in the first sub-information in described at least two first information for being calculated
The 3rd positional information of the object of identical first optimizes treatment in conjunction, obtains the 4th position of the object of the identical first
Information, as the positional information of the object of the identical first;
The first information set includes:The set of the image information frame of the smart machine collection;
Second information aggregate includes:Corresponding to the set of the depth information of each image information frame;
The first position information includes:Positional information of the smart machine in global coordinate system;
The second place information includes:Relative position information of first object relative to the smart machine;
First object includes:The characteristic point extracted from described image information frame;
First object set includes:All characteristic points in image information frame;
3rd positional information includes:Positional information of the characteristic point in global coordinate system.
2. method according to claim 1, it is characterised in that described to be set according to the second place information and the intelligence
Standby second place information in the environment calculates the first sub-information at least two first information respectively
In the first object set in each the first object the 3rd positional information, including:
By in first object set in the first sub-information in the first information in each described first information set
Spatially position is divided at least two part to all first sets;
Each in the first object set in the first sub-information at least two first information
The second place information and the smart machine of the first object first position information in the environment, described in parallel computation at least
The 3rd of each the first object in the first object set in the first sub-information in more than two first information
Confidence ceases.
3. method according to claim 1, it is characterised in that it is described to calculated it is described at least two described in
The 3rd positional information of the object of identical first is carried out in first object set in the first sub-information in the first information
Optimization processing, including:
It is the object of identical first in the first sub-information in described at least two first information for being calculated
The 3rd positional information set up Gauss model;
Determine the Gaussian distribution feature of the 3rd positional information of the object of the identical first;
The Gaussian distribution feature of the 3rd positional information according to the object of the identical first screens out variance beyond given threshold
Beyond the 3rd positional information of the object of the identical first;
Calculate the average value of the 3rd positional information of the remaining object of the identical first.
4. method according to claim 1, it is characterised in that each first information correspondence in the calculating first information set
Smart machine first position information in the environment, including:
Respectively to the smart machine move horizontally direction and vertically move direction enter row distance correction;
It is determined that obtaining the time information of all first information in first information set;
According to the first position information before smart machine distance correction, determine time information, move horizontally direction and vertical
The correction distance of straight moving direction, calculating smart machine moment of all first information in first information set is obtained
First position information.
5. a kind of smart machine;Characterized in that, the smart machine includes:First acquisition unit, the first computing unit, second
Acquiring unit, the 3rd acquiring unit, the second computing unit, the 3rd computing unit and optimization processing unit, wherein:
First acquisition unit, for obtaining first information set and the second information aggregate corresponding to the first information set;
First computing unit, for calculate obtain first information set in each first information when the corresponding smart machine in ring
First position information in border;
Second acquisition unit, for obtaining respectively from least two first information in the first information set
One sub-information;
3rd acquiring unit, for being obtained respectively from the first sub-information at least two first information
First object set;
Second computing unit, in basis second information aggregate corresponding with the first information in first information set
Second information, calculates the first object set in the first sub-information at least two first information respectively
In each the first object second place information;
3rd computing unit, according to the second place information and the smart machine second place information difference in the environment
Calculate in the first object set in the first sub-information at least two first information each first
3rd positional information of object;
Optimization processing unit, in the first sub-information in the described at least two first information to being calculated
First object set in the 3rd positional information of the object of identical first optimize treatment, obtain the identical
4th positional information of one object, as the positional information of the object of the identical first;
The first information set includes:The set of the image information frame of the smart machine collection;
Second information aggregate includes:Corresponding to the set of the depth information of each image information frame;
The first position information includes:Positional information of the smart machine in global coordinate system;
The second place information includes:Relative position information of first object relative to the smart machine;
First object includes:The characteristic point extracted from described image information frame;
First object set includes:All characteristic points in image information frame;
3rd positional information includes:Positional information of the characteristic point in global coordinate system.
6. smart machine according to claim 5, it is characterised in that the 3rd computing unit includes:Divide subelement
With parallel computation subelement, wherein:
Subelement is divided, for by described the in the first sub-information in the first information in each described first information set
Spatially position is divided at least two part to all first sets in one object set;
Parallel computation subelement, for the first sub-information at least two first information in
The second place information and the smart machine of each the first object in one object set first position letter in the environment
Cease, it is every in the first object set in the first sub-information at least two first information described in parallel computation
3rd positional information of individual first object.
7. smart machine according to claim 5, it is characterised in that the optimization processing unit includes:Set up subelement,
Determination subelement, screening subelement and computation subunit, wherein:
Subelement is set up, for in the first sub-information in described at least two first information for being calculated
3rd positional information of the object of identical first sets up Gauss model;
Determination subelement, the Gaussian distribution feature of the 3rd positional information for determining the object of the identical first;
Screening subelement, the Gaussian distribution feature for the 3rd positional information according to the object of the identical first screens out variance
Beyond the 3rd positional information beyond the object of the identical first of given threshold;
Computation subunit, the average value of the 3rd positional information for calculating the remaining object of the identical first.
8. smart machine according to claim 5, it is characterised in that first computing unit includes:Correction subelement,
Determination subelement and computation subunit, wherein:
Correction subelement, for respectively to the smart machine move horizontally direction and vertically move direction enter row distance rectify
Just;
Determination subelement, the time information for determining all first information in acquisition first information set;
Computation subunit, for the first position information according to the smart machine before correction, determines time information, water
Flat moving direction and the correction distance in direction is vertically moved, calculate the smart machine all the in first information set is obtained
The first position information at the moment of one information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310392816.9A CN104424637B (en) | 2013-09-02 | 2013-09-02 | The method and smart machine of information processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310392816.9A CN104424637B (en) | 2013-09-02 | 2013-09-02 | The method and smart machine of information processing |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104424637A CN104424637A (en) | 2015-03-18 |
CN104424637B true CN104424637B (en) | 2017-06-27 |
Family
ID=52973524
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310392816.9A Active CN104424637B (en) | 2013-09-02 | 2013-09-02 | The method and smart machine of information processing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104424637B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105957006A (en) * | 2016-04-28 | 2016-09-21 | 乐视控股(北京)有限公司 | Image processing method and device |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103252778A (en) * | 2011-12-23 | 2013-08-21 | 三星电子株式会社 | Apparatus for estimating the robot pose and method thereof |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8866889B2 (en) * | 2010-11-03 | 2014-10-21 | Microsoft Corporation | In-home depth camera calibration |
-
2013
- 2013-09-02 CN CN201310392816.9A patent/CN104424637B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103252778A (en) * | 2011-12-23 | 2013-08-21 | 三星电子株式会社 | Apparatus for estimating the robot pose and method thereof |
Non-Patent Citations (2)
Title |
---|
基于Kinect 系统的场景建模与机器人自主导航;杨东方等;《机器人》;20120930;第34卷(第5期);全文 * |
基于图优化的同时定位与地图创建综述;梁明杰等;《机器人》;20130731;第35卷(第4期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN104424637A (en) | 2015-03-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109255813B (en) | Man-machine cooperation oriented hand-held object pose real-time detection method | |
CN110026987B (en) | Method, device and equipment for generating grabbing track of mechanical arm and storage medium | |
CA2825834C (en) | Automated frame of reference calibration for augmented reality | |
CN107990899A (en) | A kind of localization method and system based on SLAM | |
Duncan et al. | Multi-scale superquadric fitting for efficient shape and pose recovery of unknown objects | |
CN107590836A (en) | A kind of charging pile Dynamic Recognition based on Kinect and localization method and system | |
CN106778790A (en) | A kind of target identification based on three-dimensional point cloud and localization method and system | |
CN108225345A (en) | The pose of movable equipment determines method, environmental modeling method and device | |
CN104732587A (en) | Depth sensor-based method of establishing indoor 3D (three-dimensional) semantic map | |
CN106846485A (en) | A kind of indoor three-dimensional modeling method and device | |
Kudoh et al. | Painting robot with multi-fingered hands and stereo vision | |
CN102853830A (en) | Robot vision navigation method based on general object recognition | |
CN112927303B (en) | Lane line-based automatic driving vehicle-mounted camera pose estimation method and system | |
CN106569591A (en) | Tracking method and system based on computer vision tracking and sensor tracking | |
CN107782304A (en) | The localization method and device of mobile robot, mobile robot and storage medium | |
CN107564065A (en) | The measuring method of man-machine minimum range under a kind of Collaborative environment | |
CN112288815A (en) | Target mode position measuring method, system, storage medium and equipment | |
CN111178170B (en) | Gesture recognition method and electronic equipment | |
Yu et al. | Accurate and robust visual localization system in large-scale appearance-changing environments | |
CN104424637B (en) | The method and smart machine of information processing | |
Han et al. | Visual tracking of a moving target using active contour based SSD algorithm | |
CN107345814A (en) | A kind of mobile robot visual alignment system and localization method | |
CN104424297B (en) | The method and smart machine of information processing | |
CN111208785B (en) | Motion control method and device | |
CN109901589A (en) | Mobile robot control method and apparatus |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |