CN109712215A - A kind of wire frame drawing drawing method and system based on robot - Google Patents
A kind of wire frame drawing drawing method and system based on robot Download PDFInfo
- Publication number
- CN109712215A CN109712215A CN201811378155.3A CN201811378155A CN109712215A CN 109712215 A CN109712215 A CN 109712215A CN 201811378155 A CN201811378155 A CN 201811378155A CN 109712215 A CN109712215 A CN 109712215A
- Authority
- CN
- China
- Prior art keywords
- lines
- face
- robot
- transformation
- image
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to a kind of wire frame drawing drawing method and system based on robot, method is the following steps are included: obtain the simple lines sequence on shooting image；Spatial relationship transformation is carried out to letter lines sequence according to preset spatial relationship transformation equation, obtains transformation lines sequence corresponding to robot coordinate system；Line frame graph corresponding with shooting image is drawn according to transformation lines sequence in robot coordinate system by robot.Wire frame drawing drawing method and system provided by the invention based on robot, robot can continuously draw line frame graph for a long time, software is extracted compared to traditional artificial drafting and line frame graph, automatic drafting line frame graph that can be more stable substantially increases the drafting efficiency of line frame graph.
Description
Technical field
The present invention relates to robotic technology field more particularly to a kind of wire frame drawing drawing method based on robot and it is
System.
Background technique
Line frame graph is the schematic diagram for describing things with brief lines, is played an important role in some application fields, such as: needle
Product design to field of industrial production, the webpage design for internet area and the personal portrait for fine arts field.
Currently, main corresponding with shooting image by manually drawing line frame graph or extracting software extraction by line frame graph
Line frame graph, the relatively slow and different line frame graphs for same target of the artificial speed for drawing line frame graph have biggish differentiation, tool
Have a unstability, line frame graph extracts software and needs user's multi-pass operation user interface that can just obtain line frame graph, the degree of automation compared with
It is low.
Summary of the invention
The technical problem to be solved by the present invention is to for the stability difference and automation for drawing line frame graph in the prior art
The lower deficiency of degree provides a kind of wire frame drawing drawing method and system based on robot.
The technical scheme to solve the above technical problems is that
According to the present invention in a first aspect, providing a kind of wire frame drawing drawing method based on robot, including following step
It is rapid:
Step 100 obtains the simple lines sequence shot on image；
Step 200 carries out spatial relationship transformation to the letter lines sequence according to preset spatial relationship transformation equation,
Obtain transformation lines sequence corresponding to robot coordinate system；
Step 300, robot are drawn in the robot coordinate system and the shooting according to the transformation lines sequence
The corresponding line frame graph of image.
Compared with the prior art, a kind of wire frame drawing drawing method bring beneficial effect based on robot of the invention
It is: by spatial relationship transformation equation by simple lines sequence physical space reachable from Image space transformation to robot, obtains
Line frame graph is drawn according to transformation lines sequence by transformation lines sequence under to robot coordinate system, robot, can continuously quickly
Line frame graph is drawn, extracts software, the automatic drafting wire frame that robot can be more stable compared to traditional artificial and line frame graph
Figure, is greatly improved the drafting efficiency of line frame graph.
Based on the above technical solution, it can make and further obtain following technical solution:
Further, the step 100 specifically includes:
Step 110 obtains shooting image, and the shooting image includes face area and in addition to the face area
Non- face area；
Step 120 carries out characteristic point detection to the shooting image using facial feature detection model, obtains the face
The location information of facial feature points and the facial feature points on the shooting image in region；
Step 130, the integrality that the human face region is judged according to the facial feature points, if the human face region is endless
It is whole, then it is back to step 110, if the human face region is complete, continues to execute step 140；
Step 140 determines the simple lines of face according to the facial feature points and the location information, and application image mentions
Method is taken to extract the image lines figure in the non-face area from the shooting image；
Step 150, the simple lines sequence is generated according to the simple lines of the face and described image lines figure.
Above-mentioned further technical solution bring has the technical effect that quickly and accurately to be examined by facial feature detection model
Facial feature points and location information are surveyed, the integrality of human face region is judged by facial feature points, when human face region is imperfect
Facial image is reacquired, the accuracy of facial feature points has been effectively ensured；When human face region is complete, by image zooming-out side
Method extracts the image lines figure in non-face area from shooting image, and realization divides image lines figure and facial feature points
Extraction is opened, ensure that image lines figure is clear.
Further, in the step 130, judge that the integrality of the human face region is specifically wrapped according to the facial feature points
It includes:
If the facial feature points include all characteristic points in face local organs, determine that the human face region is complete
It is whole, otherwise, it is determined that the human face region is imperfect.
Whether above-mentioned further technical solution bring has the technical effect that by including face office in matching facial feature points
All characteristic points on portion's organ, judge the integrality of human face region, improve the judgment accuracy of human face region.
Further, in the step 140, the simple lines of face are determined according to the facial feature points and the location information
It specifically includes:
The face local organs of the face area are positioned on the shooting image according to the positional information；
Using scheduled feature point interpolation method respectively to the adjacent face characteristic in the face local organs
Point carries out characteristic point interpolation, obtains face interpolation characteristic point；
All human face characteristic points and all face interpolation features are clicked through using scheduled lines smoothing method
Row is smoothly connected, and obtains the simple lines of the face.
Above-mentioned further technical solution bring has the technical effect that through location information quickly and accurately in shooting image
On orient face local organs, by feature point interpolation method to feature point interpolation method carry out characteristic point interpolation processing, and
And facial feature points are smoothed by lines smoothing method, so that face's letter lines continuously smooth.
Further, the simple lines of the face include that the simple lines of face and face contour lines, the step 150 are specific
Include:
According to scheduled lines storage order, by the simple lines of face and face contour line in the simple lines of the face
Item is sequentially stored in lines memory；
It is discrete to described image lines figure progress lines using scheduled lines discrete method, obtain multiple discrete lines；
According to the lines storage order, multiple discrete lines are sequentially stored in the lines memory；
The simple lines of the face, the face contour lines and multiple discrete lines are in the lines memory
Successively storage forms the simple lines sequence.
Above-mentioned further technical solution bring has the technical effect that is by image lines figure is discrete by lines discrete method
Multiple discrete lines enhance the dispersion of discrete lines, and successively orderly store multiple discrete lines by lines memory
The simple lines of item and face, are convenient for robot drawing line frame graph.
Further, in the step 200, the spatial relationship transformation equation is indicated are as follows:
Wherein, X, Y and Z represent coordinate corresponding with three vertical each other axis of orientations, T representation transformation in world coordinate system
Relational matrix, coordinate corresponding with both direction axis vertical each other in u and v representative image coordinate system；
The transformation relation matrix T is indicated are as follows:
Wherein, x and y represents the translational movement of the letter lines sequence corresponding flat transformation, and α represents the simple lines sequence
Arrange the zoom factor of corresponding change of scale.
Above-mentioned further technical solution bring has the technical effect that will be shot on image by spatial relationship transformation equation
Transformation lines matrix under the corresponding simple lines sequence transformation to robot coordinate system in plane of delineation space, and pass through transformation
Relational matrix carries out plane transformation and change of scale to letter lines sequence, and plane of delineation space is displaced by realization with suitable flat
Within the scope of spatial scaling to the drafting paper pre-set, facilitate robot drawing line frame graph.
Further, the step 300 specifically includes:
Applied robot's inverse kinematics numerical method solves the transformation lines sequence, obtains motion path；
End effector in the robot is successively drawn in the transformation lines sequence along the motion path
Each lines obtain the line frame graph.
Above-mentioned further technical solution bring has the technical effect that through the end effector in robot along moving road
Diameter successively draws each lines in transformation lines sequence, can draw out to continuously smooth each lines, end effector
Operational efficiency is high and the drafting time is long, can be realized simulation human arm and draws line frame graph, line frame graph has preferable clarity.
Second aspect according to the present invention provides a kind of line frame graph drawing system based on robot, comprising: simple line
Retrieval module, transformation lines retrieval module and robot；
The letter lines retrieval module, for obtaining the simple lines sequence on shooting image；
The transformation lines retrieval module is used for according to preset spatial relationship transformation equation to the simple lines
Sequence carries out spatial relationship transformation, obtains transformation lines sequence corresponding to robot coordinate system；
The robot, for being drawn in the robot coordinate system according to the transformation lines sequence and the shooting
The corresponding line frame graph of image.
Further, the simple lines retrieval module is specifically used for:
Shooting image is obtained, the shooting image includes face area and the non-facial regions in addition to the face area
Domain；
Characteristic point detection is carried out to the shooting image using facial feature detection model, is obtained in the face area
The location information of facial feature points and the facial feature points on the shooting image；
The integrality of the human face region is judged according to the facial feature points, if the human face region is imperfect, is weighed
Newly obtain the shooting image；
If the face area is complete, the simple line of face is determined according to the facial feature points and the location information
Item, and application image extracting method extracts the image lines figure in the non-face area from the shooting image；
The simple lines sequence is generated according to the simple lines of the face and described image lines figure.
Further, the spatial relationship transformation equation indicates are as follows:
Wherein, X, Y and Z represent coordinate corresponding with three vertical each other axis of orientations, T representation transformation in world coordinate system
Relational matrix, coordinate corresponding with both direction axis vertical each other in u and v representative image coordinate system；
The transformation relation matrix T is indicated are as follows:
Wherein, x and y represents the translational movement of the letter lines sequence corresponding flat transformation, and α represents the simple lines sequence
Arrange the zoom factor of corresponding change of scale.
Compared with the prior art, a kind of wire frame drawing drawing method bring beneficial effect based on robot of the invention
Be: transformation lines retrieval module by spatial relationship transformation equation by simple lines sequence from Image space transformation to machine
The reachable physical space of people, obtains the transformation lines sequence under robot coordinate system, and robot is drawn according to transformation lines sequence
Line frame graph, compared to traditional artificial and line frame graph extraction software, the automatic drafting line frame graph that robot can be more stable, greatly
Amplitude improves the drafting efficiency of line frame graph, and can be realized Fast Drawing line frame graph corresponding with any shooting image, line
Block diagram has many advantages, such as that the lines flow smoothly, clear and higher with shooting image similarity.
Detailed description of the invention
Fig. 1 is a kind of flow diagram for wire frame drawing drawing method based on robot that the embodiment of the present invention one provides；
Fig. 2 is the flow diagram corresponding to step 100 in Fig. 1；
The flow diagram of image lines figure is extracted in the slave facial image that Fig. 3 provides for the embodiment of the present invention one；
Fig. 4 is the flow diagram that the slave facial image that the embodiment of the present invention one provides extracts the simple lines of face；
Fig. 5 is a kind of structural schematic diagram of the line frame graph drawing system based on robot provided by Embodiment 2 of the present invention；
Fig. 6 is the structural representation of another line frame graph drawing system based on robot provided by Embodiment 2 of the present invention
Figure；
Fig. 7 is a kind of structural schematic diagram for wire frame drawing system based on robot that the embodiment of the present invention three provides.
In attached drawing, parts list represented by the reference numerals are as follows:
1- high definition camera, 2- host computer, 3- slave computer, 4- robot, 5- end effector.
Specific embodiment
The principle and features of the present invention will be described below with reference to the accompanying drawings, and the given examples are served only to explain the present invention, and
It is non-to be used to limit the scope of the invention.
Embodiment one
As shown in Figure 1, a kind of flow diagram of wire frame drawing drawing method based on robot of the embodiment of the present invention, this
Embodiment is using facial image as shooting image, the progress example description using portrait schematic diagram as line frame graph, comprising the following steps:
Step 100 obtains the simple lines sequence shot on image；
Step 200 carries out spatial relationship transformation to letter lines sequence according to preset spatial relationship transformation equation, obtains
Transformation lines sequence corresponding to robot coordinate system；
Line corresponding with shooting image is drawn according to transformation lines sequence in robot coordinate system by step 300, robot
Block diagram.
It is by spatial relationship transformation equation that simple lines sequence physics reachable from Image space transformation to robot is empty
Between, the transformation lines sequence under robot coordinate system is obtained, line frame graph is drawn according to transformation lines sequence by robot, can be continuous
Fast Drawing line frame graph extracts software, the automatic drafting that robot can be more stable compared to traditional artificial and line frame graph
Line frame graph, is greatly improved the drafting efficiency of line frame graph, and it is corresponding with any shooting image to can be realized Fast Drawing
Line frame graph, line frame graph have many advantages, such as that the lines flow smoothly, clear and higher with shooting image similarity.
Preferably, as shown in Fig. 2, step 100 specifically includes:
Step 110 obtains shooting image, and shooting image includes face area and the non-facial regions in addition to face area
Domain；
Step 120 carries out characteristic point detection to shooting image using facial feature detection model, obtains in face area
The location information of facial feature points and facial feature points in shooting image；
Step 130, the integrality that human face region is judged according to facial feature points return to step if human face region is imperfect
Rapid 110, if human face region is complete, continue to execute step 140；
Step 140 determines the simple lines of face according to facial feature points and location information, and application image extracting method from
The image lines figure in non-face area is extracted on shooting image；
Step 150, simple lines sequence is generated according to the simple lines of face and image lines figure.
Facial feature detection model carries out what learning training obtained to a large amount of human face datas using pre-training model, pre-training
Model is formed according to the shape description feature of face local organs and apart from Expressive Features, and face local organs include eyebrow
Hair, eyes, nose, mouth and face contour, human face data include Static Human Face image, dynamic human face image, have different faces
The facial image in region and the facial image with different faces expression etc..
Facial feature points include 64 characteristic points on eyebrow, eyes, nose, mouth and face contour, and location information includes
It is numbered with the one-to-one characteristic point of each characteristic point in 64 characteristic points, each characteristic point number indicates characteristic point in face
Relative position on image.
Image extraction method includes image filtering, Image Edge-Detection, morphological image operation and the methods of connected domain, such as
Fig. 3 show the flow diagram that image lines figure is extracted in the slave facial image of the offer of the embodiment of the present invention one.
Facial feature points and location information are quickly and accurately detected by facial feature detection model, pass through face feature
Point judges the integrality of human face region, and facial image is reacquired when human face region is imperfect, face feature has been effectively ensured
The accuracy of point；When human face region is complete, the figure in non-face area is extracted from shooting image by image extraction method
As lines figure, realizes and image lines figure and facial feature points are carried out to separate extraction, ensure that image lines figure is clear.
Preferably, in step 130, judge that the integrality of human face region specifically includes according to facial feature points: if face is special
Sign point includes all characteristic points in face local organs, then determines that human face region is complete, otherwise, it is determined that human face region is endless
It is whole.
By, whether comprising all characteristic points in face local organs, judging human face region in matching facial feature points
Integrality improves the judgment accuracy of human face region.
Preferably, in step 140, determine that the simple lines of face specifically include according to facial feature points and location information: root
Position the face local organs of face area in shooting image according to location information；Distinguish using scheduled feature point interpolation method
Feature point interpolation is carried out to the adjacent human face characteristic point in face local organs, obtains face interpolation characteristic point；Using predetermined
Lines smoothing method all human face characteristic points and all face interpolation characteristic points are smoothly connected, obtain the simple line of face
Item.
Feature point interpolation method is the characteristic point that predetermined number is inserted between the adjacent characteristic point in face local organs,
Predetermined number can be 1 or 2, and lines smoothing method can use average smooth bus connection method, be illustrated in figure 4 this hair
The slave facial image that bright embodiment one provides extracts the flow diagram of the simple lines of face.
Face local organs are quickly and accurately oriented in shooting image by location information, pass through feature point interpolation
Method carries out characteristic point interpolation processing to feature point interpolation method, and is carried out to facial feature points by lines smoothing method flat
Sliding processing, so that face's letter lines continuously smooth.
Preferably, the simple lines of face include that the simple lines of face and face contour lines, step 150 specifically include: being pressed
According to scheduled lines storage order, by the simple lines of face the simple lines of face and face contour lines be sequentially stored in line
In memory；It is discrete to image lines figure progress lines using scheduled lines discrete method, obtain multiple discrete lines；It presses
According to lines storage order, multiple discrete lines are sequentially stored in lines memory；Face letter lines, face contour lines
Simple lines sequence is formed with successively storing in multiple online memories of discrete lines.
Lines storage order can be the sequence of facial image from top to bottom, from outside to inside, or from inside to outside, from it is lower to
On sequence；Lines discrete method can use connected domain method, for the simple lines of each of image lines figure are discrete
The discrete lines drawn at one, such as: a letter lines with small acute angle or bifurcated are broken as two discrete lines
Item；Multiple simple lines in the simple lines of multiple discrete lines and face are stored in lines memory in the form of data group
In, and multiple discrete lines can be in the storage location after the simple lines of face.
By lines discrete method by image lines figure it is discrete be multiple discrete lines, enhance the discrete of discrete lines
Degree, and the simple lines of multiple discrete lines and face are successively orderly stored by lines memory, it is convenient for robot drawing line
Block diagram.
Preferably, in step 200, spatial relationship transformation equation indicates are as follows:
Wherein, X, Y and Z represent coordinate corresponding with three vertical each other axis of orientations, T representation transformation in world coordinate system
Relational matrix, coordinate corresponding with both direction axis vertical each other in u and v representative image coordinate system.
Transformation relation matrix T is indicated are as follows:
Wherein, x and y represents the translational movement of letter lines sequence corresponding flat transformation, and α represents a simple lines sequence and corresponds to ruler
Spend the zoom factor of transformation.
The corresponding simple lines sequence transformation in the plane of delineation space on image will be shot by spatial relationship transformation equation
Transformation lines matrix under to robot coordinate system, and plane transformation is carried out to letter lines sequence by transformation relation matrix
And change of scale, it realizes plane of delineation space with suitable flat displacement and spatial scaling to the drafting paper model pre-set
In enclosing, facilitate robot drawing line frame graph.
Preferably, step 300 specifically includes: applied robot's inverse kinematics numerical method seeks transformation lines sequence
Solution, obtains motion path；End effector in robot successively draws each of transformation lines sequence along motion path
Lines obtain the line frame graph.
Robot can be six-joint robot, and six-joint robot includes end effector, and end effector includes multiple passes
Section；End effector successively rotates angle according to each path point in motion path, to reach corresponding predetermined position, in phase
It is smoothly moved between adjacent path point, until having traversed all path points, realizes and draw line frame graph.
The each lines converted in lines sequence are successively drawn along motion path by the end effector in robot,
Each lines can be drawn out to continuously smooth, the operational efficiency of end effector is high and the drafting time is long, can be realized simulation
Human arm draws line frame graph, and line frame graph has preferable clarity.
Embodiment two
As shown in figure 5, a kind of structural schematic diagram of line frame graph drawing system based on robot of the embodiment of the present invention, packet
It includes: simple lines retrieval module, transformation lines retrieval module and robot；Simple lines retrieval module, is used
Simple lines sequence on acquisition shooting image；Lines retrieval module is converted, for becoming according to preset spatial relationship
It changes equation and spatial relationship transformation is carried out to letter lines sequence, obtain transformation lines sequence corresponding to robot coordinate system；Machine
Device people, for drawing line frame graph corresponding with shooting image in robot coordinate system according to transformation lines sequence.
Preferably, a simple lines retrieval module is specifically used for: obtaining shooting image, shooting image includes facial regions
Domain and the non-face area in addition to face area；Characteristic point detection is carried out to shooting image using facial feature detection model,
Obtain the location information of facial feature points and facial feature points in shooting image in face area；According to facial feature points
Judge the integrality of human face region, if human face region is imperfect, reacquires shooting image；If face area is complete, root
The simple lines of face are determined according to facial feature points and location information, and application image extracting method extracts non-face from shooting image
Image lines figure in portion region；Simple lines sequence is generated according to the simple lines of face and image lines figure.
Preferably, a simple lines retrieval module is specifically used for: if facial feature points include in face local organs
All characteristic points, then determine that human face region is complete, otherwise, it is determined that human face region is imperfect.
Preferably, a simple lines retrieval module is specifically used for: positioning face in shooting image according to location information
The face local organs in region；It is special to the adjacent face in face local organs respectively using scheduled feature point interpolation method
Sign point carries out characteristic point interpolation, obtains face interpolation characteristic point；Using scheduled lines smoothing method to all human face characteristic points
It is smoothly connected with all face interpolation characteristic points, obtains the simple lines of face.
Preferably, as shown in fig. 6, further including lines memory, face's letter lines include the simple lines of face and shape of face
Lines of outline, a simple lines retrieval module are specifically used for:, will be in the simple lines of face according to scheduled lines storage order
The simple lines of face and face contour lines be sequentially stored in lines memory；Using scheduled lines discrete method to figure
Picture lines figure progress lines are discrete, obtain multiple discrete lines；According to lines storage order, multiple discrete lines are sequentially stored
In lines memory；It is successively stored in online face letter lines, face contour lines and multiple discrete lines memory
Form simple lines sequence.
Preferably, transformation lines retrieval module is stored with spatial relationship transformation equation and transformation relation matrix, space
Relation transformation equation indicates are as follows:
Wherein, X, Y and Z represent coordinate corresponding with three vertical each other axis of orientations, T representation transformation in world coordinate system
Relational matrix, coordinate corresponding with both direction axis vertical each other in u and v representative image coordinate system.
Transformation relation matrix T is indicated are as follows:
Wherein, x and y represents the translational movement of letter lines sequence corresponding flat transformation, and α represents a simple lines sequence and corresponds to ruler
Spend the zoom factor of transformation.
Preferably, as shown in fig. 6, further including that motion path obtains module, motion path obtains module and is used to apply machine
People's inverse kinematics numerical method solves the transformation lines sequence, obtains motion path；Control the end in robot
Actuator is along movement path；End effector in robot is successively drawn along motion path in transformation lines sequence
Each lines, obtain line frame graph.
A kind of line frame graph drawing system based on robot that the present embodiment two provides, transformation lines retrieval module are logical
Spatial relationship transformation equation is crossed by simple lines sequence physical space reachable from Image space transformation to robot, obtains machine
Line frame graph is drawn according to transformation lines sequence by transformation lines sequence under people's coordinate system, robot, compared to traditional artificial and
Line frame graph extracts software, and the drafting effect of line frame graph is greatly improved in the automatic drafting line frame graph that robot can be more stable
Rate, and can be realized Fast Drawing line frame graph corresponding with any shooting image, line frame graph have the lines flow smoothly, it is clear and with
Shoot the advantages that image similarity is higher.
Embodiment three
As shown in fig. 7, a kind of structural schematic diagram of line frame graph drawing system based on robot of the embodiment of the present invention, packet
Include high definition camera, host computer, slave computer and robot.
High definition camera, for acquiring shooting image and sending shooting image to host computer.
Host computer, for receiving shooting image；Obtain the simple lines sequence on shooting image；It is closed according to preset space
It is that transformation equation carries out spatial relationship transformation to letter lines sequence, obtains transformation lines sequence corresponding to robot coordinate system
Column；Transformation lines sequence synchronization is sent to slave computer and robot；
Slave computer solves transformation lines sequence for applied robot's inverse kinematics numerical method, is moved
Path；It is sent according to motion path and draws instruction to robot；
Instruction is drawn for receiving by robot；It is instructed according to drawing, controls the end effector in robot along moving
Each lines in transformation lines sequence are successively drawn in path in robot coordinate system, obtain line frame graph.
Before acquisition shoots image, high definition camera, host computer, slave computer and robot are mounted on station, it is high
Clear camera is located at the side of station, facilitates acquisition shooting image；Host computer communicates electricity with slave computer and robot respectively
Cable connection is transferred to slave computer and robot to realize host computer for lines sequence synchronization is converted；Slave computer passes through communication cable
It is connect with robot, for controlling end effector according to movement path.
When initializing high definition camera, focal length, resolution ratio, the window for adjusting high definition camera such as show at the parameters, so that shooting
The face feature that image is more clear, shoots on image is more obvious, facilitates robot drawing line frame graph；Robot can be six
Axis robot obtains transformation relation matrix by six axis machines of initialization.
Preferably, host computer is specifically used for: carrying out characteristic point detection to shooting image using facial feature detection model, obtains
The location information of facial feature points and facial feature points in shooting image in face area；Sentenced according to facial feature points
The integrality of disconnected human face region receives shooting image if human face region is imperfect again；If face area is complete, basis
Facial feature points and location information determine the simple lines of face, and application image extracting method extracts non-face from shooting image
Image lines figure in region；Simple lines sequence is generated according to the simple lines of face and image lines figure.
Preferably, host computer is specifically used for: if facial feature points include all characteristic points in face local organs,
Determine that human face region is complete, otherwise, it is determined that human face region is imperfect.
Preferably, host computer is specifically used for: positioning the face part of face area in shooting image according to location information
Organ；Feature point interpolation is carried out to all adjacent human face characteristic points in face local organs using feature point interpolation method,
Obtain face interpolation characteristic point；Using scheduled lines smoothing method to all human face characteristic points and all face interpolation characteristic points
It is smoothly connected, obtains the simple lines of face.
Preferably, host computer is specifically used for: according to scheduled lines storage order, by the face letter in the simple lines of face
Lines and face contour lines are sequentially stored in lines memory；Using scheduled lines discrete method to image lines figure
It is discrete to carry out lines, obtains multiple discrete lines；According to lines storage order, multiple discrete lines are sequentially stored in lines and are deposited
In reservoir；Successively storage forms simple pen in online face letter lines, face contour lines and multiple discrete lines memory
Lines sequence.
Preferably, host computer is stored with spatial relationship transformation equation and transformation relation matrix, spatial relationship transformation equation table
It is shown as:
Wherein, X, Y and Z represent coordinate corresponding with three vertical each other axis of orientations, T representation transformation in world coordinate system
Relational matrix, coordinate corresponding with both direction axis vertical each other in u and v representative image coordinate system.
Transformation relation matrix T is indicated are as follows:
Wherein, x and y represents the translational movement of letter lines sequence corresponding flat transformation, and α represents a simple lines sequence and corresponds to ruler
Spend the zoom factor of transformation.
A kind of line frame graph drawing system based on robot that the present embodiment three provides, host computer are converted by spatial relationship
Simple lines sequence physical space reachable from Image space transformation to robot is obtained the change under robot coordinate system by equation
Line frame graph is drawn according to transformation lines sequence by thread-changing sequence, robot, extracts software compared to traditional artificial and line frame graph,
Robot can be more stable automatic drafting line frame graph, the drafting efficiency of line frame graph is greatly improved, can be realized quickly
Draw line frame graph corresponding with any shooting image, line frame graph has that the lines flow smoothly, clear and higher with image similarity is shot
The advantages that.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of wire frame drawing drawing method based on robot, which comprises the following steps:
Step 100 obtains the simple lines sequence shot on image；
Step 200 carries out spatial relationship transformation to the letter lines sequence according to preset spatial relationship transformation equation, obtains
Transformation lines sequence corresponding to robot coordinate system；
Step 300, robot are drawn in the robot coordinate system and the shooting image according to the transformation lines sequence
Corresponding line frame graph.
2. a kind of portrait schematic diagram method for drafting based on robot according to claim 1, which is characterized in that the step
100 specifically include:
Step 110 obtains shooting image, and the shooting image includes face area and the non-face in addition to the face area
Portion region；
Step 120 carries out characteristic point detection to the shooting image using facial feature detection model, obtains the face area
The location information of interior facial feature points and the facial feature points on the shooting image；
Step 130, the integrality that the human face region is judged according to the facial feature points, if the human face region is imperfect,
It is then back to step 110, if the human face region is complete, continues to execute step 140；
Step 140 determines the simple lines of face, and application image extraction side according to the facial feature points and the location information
Method extracts the image lines figure in the non-face area from the shooting image；
Step 150, the simple lines sequence is generated according to the simple lines of the face and described image lines figure.
3. a kind of wire frame drawing drawing method based on robot according to claim 2, which is characterized in that the step
In 130, judge that the integrality of the human face region specifically includes according to the facial feature points:
If the facial feature points include all characteristic points in face local organs, determine that the human face region is complete,
Otherwise, it is determined that the human face region is imperfect.
4. a kind of wire frame drawing drawing method based on robot according to claim 2, which is characterized in that the step
In 140, determine that the simple lines of face specifically include according to the facial feature points and the location information:
The face local organs of the face area are positioned on the shooting image according to the positional information；
The adjacent face characteristic in the face local organs is clicked through respectively using scheduled feature point interpolation method
Row feature point interpolation obtains face interpolation characteristic point；
All human face characteristic points and all face interpolation characteristic points are carried out using scheduled lines smoothing method flat
Slip obtains the simple lines of the face.
5. a kind of wire frame drawing drawing method based on robot according to claim 2, which is characterized in that face's letter
Lines include that the simple lines of face and face contour lines, the step 150 specifically include:
According to scheduled lines storage order, the simple lines of the face and face contour lines are sequentially stored in lines storage
In device；
It is discrete to described image lines figure progress lines using scheduled lines discrete method, obtain multiple discrete lines；
According to the lines storage order, multiple discrete lines are sequentially stored in the lines memory；
The simple lines of the face, the face contour lines and multiple discrete lines in the lines memory successively
Storage forms the simple lines sequence.
6. a kind of wire frame drawing drawing method based on robot according to claim 1-5, which is characterized in that
In the step 200, the spatial relationship transformation equation is indicated are as follows:
Wherein, X, Y and Z represent coordinate corresponding with three vertical each other axis of orientations, T representation transformation relationship in world coordinate system
Matrix, coordinate corresponding with both direction axis vertical each other in u and v representative image coordinate system；
The transformation relation matrix T is indicated are as follows:
Wherein, x and y represents the translational movement of the letter lines sequence corresponding flat transformation, and α represents the simple lines sequence pair
Answer the zoom factor of change of scale.
7. a kind of wire frame drawing drawing method based on robot according to claim 1-5, which is characterized in that institute
Step 300 is stated to specifically include:
Applied robot's inverse kinematics numerical method solves the transformation lines sequence, obtains motion path；
End effector in the robot successively draws each of described transformation lines sequence along the motion path
Lines obtain the line frame graph.
8. a kind of line frame graph drawing system based on robot characterized by comprising simple lines retrieval module becomes
Thread-changing retrieval module and robot；
The letter lines retrieval module, for obtaining the simple lines sequence on shooting image；
The transformation lines retrieval module is used for according to preset spatial relationship transformation equation to the simple lines sequence
Spatial relationship transformation is carried out, transformation lines sequence corresponding to robot coordinate system is obtained；
The robot, for being drawn in the robot coordinate system according to the transformation lines sequence and the shooting image
Corresponding line frame graph.
9. a kind of line frame graph drawing system based on robot according to claim 8, which is characterized in that the letter line
Retrieval module is specifically used for:
Shooting image is obtained, the shooting image includes face area and the non-face area in addition to the face area；
Characteristic point detection is carried out to the shooting image using facial feature detection model, obtains the face in the face area
The location information of characteristic point and the facial feature points on the shooting image；
The integrality of the human face region is judged according to the facial feature points, if the human face region is imperfect, is obtained again
Take the shooting image；
If the face area is complete, the simple lines of face are determined according to the facial feature points and the location information, and
Application image extracting method extracts the image lines figure in the non-face area from the shooting image；
The simple lines sequence is generated according to the simple lines of the face and described image lines figure.
10. according to a kind of described in any item line frame graph drawing systems based on robot of claim 8-9, which is characterized in that
The spatial relationship transformation equation indicates are as follows:
Wherein, X, Y and Z represent coordinate corresponding with three vertical each other axis of orientations, T representation transformation relationship in world coordinate system
Matrix, coordinate corresponding with both direction axis vertical each other in u and v representative image coordinate system；
The transformation relation matrix T is indicated are as follows:
Wherein, x and y represents the translational movement of the letter lines sequence corresponding flat transformation, and α represents the simple lines sequence pair
Answer the zoom factor of change of scale.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811378155.3A CN109712215B (en) | 2018-11-19 | 2018-11-19 | Robot-based line block diagram drawing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811378155.3A CN109712215B (en) | 2018-11-19 | 2018-11-19 | Robot-based line block diagram drawing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109712215A true CN109712215A (en) | 2019-05-03 |
| CN109712215B CN109712215B (en) | 2023-06-23 |
Family
ID=66254220
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811378155.3A Active CN109712215B (en) | 2018-11-19 | 2018-11-19 | Robot-based line block diagram drawing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109712215B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009048234A (en) * | 2007-08-13 | 2009-03-05 | Ｔａｋｕｍｉ Ｖｉｓｉｏｎ株式会社 | System and method for face recognition |
| CN105701437A (en) * | 2014-11-11 | 2016-06-22 | 沈阳新松机器人自动化股份有限公司 | Portrait drawing system based robot |
| CN106651988A (en) * | 2016-10-13 | 2017-05-10 | 中国科学院半导体研究所 | Automatic drawing system for face line paint |
-
2018
- 2018-11-19 CN CN201811378155.3A patent/CN109712215B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2009048234A (en) * | 2007-08-13 | 2009-03-05 | Ｔａｋｕｍｉ Ｖｉｓｉｏｎ株式会社 | System and method for face recognition |
| CN105701437A (en) * | 2014-11-11 | 2016-06-22 | 沈阳新松机器人自动化股份有限公司 | Portrait drawing system based robot |
| CN106651988A (en) * | 2016-10-13 | 2017-05-10 | 中国科学院半导体研究所 | Automatic drawing system for face line paint |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109712215B (en) | 2023-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Pavlovic et al. | Visual interpretation of hand gestures for human-computer interaction: A review | |
| Elforaici et al. | Posture recognition using an RGB-D camera: exploring 3D body modeling and deep learning approaches | |
| Papazov et al. | Real-time 3D head pose and facial landmark estimation from depth images using triangular surface patch features | |
| CN101452582B (en) | Method and device for implementing three-dimensional video specific action | |
| Park et al. | Articulated pose estimation with tiny synthetic videos | |
| CN109993073B (en) | Leap Motion-based complex dynamic gesture recognition method | |
| CN110363867A (en) | Virtual dress up system, method, equipment and medium | |
| CN105118023B (en) | Real-time video human face cartoon generation method based on human face characteristic point | |
| CN108171748A (en) | A kind of visual identity of object manipulator intelligent grabbing application and localization method | |
| CN109794948A (en) | Distribution network live line work robot and recognition positioning method | |
| Sweeney et al. | A model of shared grasp affordances from demonstration | |
| Feng et al. | Real-time fingertip tracking and detection using Kinect depth sensor for a new writing-in-the air system | |
| CN104407694A (en) | Man-machine interaction method and device combining human face and gesture control | |
| CN107688391A (en) | A kind of gesture identification method and device based on monocular vision | |
| CN110348496A (en) | A kind of method and system of facial image fusion | |
| Wang et al. | Robot calligraphy using pseudospectral optimal control in conjunction with a novel dynamic brush model | |
| CN115816460A (en) | Manipulator grabbing method based on deep learning target detection and image segmentation | |
| CN108830222A (en) | A kind of micro- expression recognition method based on informedness and representative Active Learning | |
| CN113711275A (en) | Creating training data variability for object labeling in images in machine learning | |
| CN110751097A (en) | Semi-supervised three-dimensional point cloud gesture key point detection method | |
| CN115861780B (en) | Robot arm detection grabbing method based on YOLO-GGCNN | |
| CN109087343A (en) | A kind of generation method and system of workpiece grabbing template | |
| CN109711324A (en) | Human posture recognition method based on Fourier transformation and convolutional neural networks | |
| Sun et al. | Autonomous clothes manipulation using a hierarchical vision architecture | |
| CN109712215A (en) | A kind of wire frame drawing drawing method and system based on robot |
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 |