CN107292893A - Image processing apparatus, image processing method and image processing equipment - Google Patents
Image processing apparatus, image processing method and image processing equipment Download PDFInfo
- Publication number
- CN107292893A CN107292893A CN201610197282.8A CN201610197282A CN107292893A CN 107292893 A CN107292893 A CN 107292893A CN 201610197282 A CN201610197282 A CN 201610197282A CN 107292893 A CN107292893 A CN 107292893A
- Authority
- CN
- China
- Prior art keywords
- straight line
- point
- angle
- cluster
- profile
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Abstract
Present disclose provides image processing apparatus, image processing method and image processing equipment.Image processing apparatus includes:First cluster cell, it is for each predetermined point on the profile, by it is being obtained based on the profile, by candidate's straight line of the point according to the angle of candidate's straight line cluster be one or more clusters, and the average value of the angle of candidate's straight line in each cluster in one or more of clusters is calculated respectively, using the angle as one or more selected straight lines for the point;And second cluster cell, it is clustered the predetermined point on the profile according to the angle of the selected straight line of each predetermined point and the position of the point, and by the point in resulting each cluster using in units of cluster as the contour segment extracted from the profile, using the contour feature as the pending image.
Description
Technical field
The disclosure relates generally to image processing field, in particular to for the wheel in image
Image processing apparatus, image processing method and image processing equipment that exterior feature is split.
Background technology
In image processing field, it is very important research class to carry out segmentation one to the profile in image
Topic.The result of contours segmentation can have an application widely, such as object identification, Stereo matching,
Text analyzing etc..
Because profile is typically to be made up of parts of some separation, therefore traditional method is mostly by seeking
The method of communication line is looked for realize the segmentation of profile.But, communication line still may have arbitrary shape
Shape, thus may not still include the desired effective information on profile.For example, in traditional wheel
In wide dividing method, a triangle can be considered as a communication line, thus directly by whole triangle
Shape as contours segmentation result.However, in actual applications it may be desirable to can be by the wheel of triangle
Exterior feature is divided into three sides.
It is desirable to extract more more than communication line of the prior art from image by contours segmentation
Feature, in favor of the subsequent treatment image-related with this.
The content of the invention
The brief overview on the present invention is given below, to provide on some of the present invention
The basic comprehension of aspect.It should be appreciated that this general introduction is not the exhaustive general introduction on the present invention.
It is not intended to determine the key or pith of the present invention, nor the model of the intended limitation present invention
Enclose.Its purpose only provides some concepts in simplified form, more detailed in this, as what is discussed later
The preamble carefully described.
In view of the defect of prior art, an object of the present invention is to provide one kind can be by arbitrary shape
Contours segmentation into the image processing apparatus of the contour segment based on contour shape, method and equipment, with
The problem of at least solving existing.
According to an aspect of this disclosure there is provided a kind of image processing apparatus, for based on pending figure
Contours extract contour segment as in, described device includes:First cluster cell, it is directed to the profile
On each predetermined point, by it is being obtained based on the profile, by candidate's straight line of the point according to described
The angle cluster of candidate's straight line is one or more clusters, and is calculated respectively in one or more of clusters
The average value of the angle of candidate's straight line in each cluster, using as the one or more of the point
The angle of selected straight line;And second cluster cell, it is by the predetermined point on the profile according to each
The angle of the selected straight line of predetermined point and the position of the point are clustered, and by resulting each cluster
In point using in units of cluster as the contour segment extracted from the profile, to be used as the pending figure
The contour feature of picture.
According to another aspect of the present disclosure there is provided a kind of image processing method, for based on pending figure
Contours extract contour segment as in, methods described includes:For each predetermined point on the profile,
By it is being obtained based on the profile, gathered according to the angle of candidate's straight line by candidate's straight line of the point
Class is one or more clusters, and the candidate calculated respectively in each cluster in one or more of clusters is straight
The average value of the angle of line, using the angle as one or more selected straight lines for the point;
And by the predetermined point on the profile according to the angle of the selected straight line of each predetermined point and the point
Position clustered, and by the point in resulting each cluster using in units of cluster as from the profile
The contour segment of middle extraction, using the contour feature as the pending image.
According to the another aspect of the disclosure there is provided a kind of image processing equipment, for based on pending figure
Contours extract contour segment as in, the equipment includes controller.The controller is configured as:Pin
, will be being obtained based on the profile, straight by the candidate of the point to each predetermined point on the profile
Line according to the angle of candidate's straight line cluster be one or more clusters, and calculate respectively it is one or
The average value of the angle of candidate's straight line in each cluster in multiple clusters, using as the point
The angle of one or more selected straight lines;And by the predetermined point on the profile according to each predetermined point
The angle of selected straight line and the position of the point clustered, and by the point in resulting each cluster
Using in units of cluster as the contour segment extracted from the profile, using the wheel as the pending image
Wide feature.
According to the other side of the disclosure, additionally provide a kind of so that computer is used as figure as described above
As the program of processing unit.
According to the another aspect of the disclosure, corresponding computer-readable recording medium is additionally provided, the meter
Be stored with calculation machine readable storage medium storing program for executing can by computing device computer program, the computer
Program can make the above-mentioned image processing method of computing device upon execution.
The above-mentioned various aspects according to the embodiment of the present disclosure, at least result in following benefit:Can be by
The contours segmentation of arbitrary shape in image is into the contour segment based on contour shape, in favor of utilizing segmentation
As a result subsequent treatment is carried out as the contour feature of image.
By the detailed description below in conjunction with accompanying drawing most preferred embodiment of this disclosure, the disclosure these
And other advantages will be apparent from.
Brief description of the drawings
The disclosure can be by reference to preferably being managed below in association with the description given by accompanying drawing
Solution, wherein having used same or analogous reference in all of the figs to represent same or similar
Part.The accompanying drawing includes in this manual together with following detailed description and forms this
A part for specification, and for preferred embodiment of the present disclosure is further illustrated and this is explained
Principle disclosed and advantage.Wherein:
Fig. 1 is the exemplary construction for schematically showing the image processing apparatus according to the embodiment of the present disclosure
Block diagram.
Fig. 2A and Fig. 2 B are to be carried out with Xi in explanation according to the image processing apparatus of the embodiment of the present disclosure
Example process explanation figure.
Fig. 3 is for illustrating that the first cluster in the image processing apparatus according to the embodiment of the present disclosure is single
The explanation figure for the example process that member is carried out.
Fig. 4 is another example for schematically showing the image processing apparatus according to the embodiment of the present disclosure
The block diagram of structure.
Fig. 5 A and Fig. 5 B are pre- in the image processing apparatus according to the embodiment of the present disclosure for illustrating
The explanation figure for the example process that processing unit is carried out.
Fig. 6 is the another example for schematically showing the image processing apparatus according to the embodiment of the present disclosure
The block diagram of structure.
Fig. 7 is the example flow for schematically showing the image processing method according to the embodiment of the present disclosure
Flow chart.
Fig. 8 is another example for schematically showing the image processing method according to the embodiment of the present disclosure
The flow chart of flow.
Fig. 9 is the another example for schematically showing the image processing method according to the embodiment of the present disclosure
The flow chart of flow.
Figure 10 is the exemplary construction for schematically showing the image processing equipment according to the embodiment of the present disclosure
Block diagram.
Figure 11 is to show to can be used to realize image processing apparatus according to the embodiment of the present disclosure, method
And a kind of structure diagram of possible hardware configuration of equipment.
Embodiment
The one exemplary embodiment of the present invention is described hereinafter in connection with accompanying drawing.In order to clear and
For the sake of simplicity, all features of actual embodiment are not described in the description.However, should
Solution, must make many specific to embodiment during any this practical embodiments are developed
Determine, to realize the objectives of developer, for example, meeting that related to system and business
A little restrictive conditions, and these restrictive conditions may change with the difference of embodiment.
In addition, it also should be appreciated that, although development is likely to be extremely complex and time-consuming, but to benefiting
For those skilled in the art of present disclosure, this development is only routine task.
Herein, in addition it is also necessary to which explanation is a bit, in order to avoid having obscured this hair because of unnecessary details
It is bright, illustrate only in the accompanying drawings with according to the closely related apparatus structure of the solution of the present invention and/or
Process step, and eliminate and the little other details of relation of the present invention.
In image processing field, expect to split the profile in image, to obtain the profile of image
Feature is with subsequent treatment of the utilization to the image.
Based on this, the present disclosure proposes a kind of image processing apparatus, method and equipment, it can be by
The contours segmentation of arbitrary shape in image is into the contour segment based on contour shape, in favor of utilizing segmentation
As a result subsequent treatment is carried out as the contour feature of image.
According to an aspect of this disclosure, there is provided a kind of image processing apparatus.Fig. 1 is schematically
The block diagram of the exemplary construction of image processing apparatus according to the embodiment of the present disclosure is shown.
The image processing apparatus of the embodiment of the present disclosure is used for based on the contours extract wheel in pending image
Wide section.As shown in figure 1, image processing apparatus 10 includes:First cluster cell 101, it is directed to
Each predetermined point on the profile, by candidate's straight line being obtained based on the profile, by the point
It is one or more clusters according to the angle of candidate's straight line cluster, and calculates one or many respectively
The average value of the angle of candidate's straight line in each cluster in individual cluster, using as one for the point
The angle of individual or multiple selected straight lines;And second cluster cell 102, it will be pre- on the profile
Fixed point is clustered according to the angle of selected straight line and the position of the point of each predetermined point, and by institute
Point in obtained each cluster using in units of cluster as the contour segment extracted from the profile, using as
The contour feature of the pending image.
The processing carried out by the first cluster cell 101, the profile based on image being obtained,
Clustered by angle of candidate's straight line based on these straight lines of a predetermined point on profile, so that
One or more selected straight lines by the predetermined point are obtained, on such selected line reflection profile
Line portion that may be present by the point.
Candidate's straight line by the predetermined point on profile can be obtained by various modes and is provided
To the first cluster cell 101.For example, it is possible to use the technology such as straight-line detection well known in the prior art
Or candidate's straight line is obtained by the acquiring unit 600 being described in detail later in reference to Fig. 6, do not enter herein
Row is described in detail.
In addition, as an example, the first cluster cell 101 carry out processing in, can be by profile
A little all as the predetermined point on profile, the point that will can also be selected on profile every n point
(n is predetermined natural number) is used as above-mentioned predetermined point.Those skilled in the art can be according to reality
The various factors such as application needs, calculated load, the processing speed on border suitably set predetermined on profile
Point, here without being described in detail.
The processing carried out by the second cluster cell 102, can be close by position on profile and be selected
The predetermined point cluster that the angle of boning out is close is arrived in identical cluster (that is, identical contour segment), from
And the contour segment of being split the predetermined point on image is obtained, using the contour feature as image.
Fig. 2A and Fig. 2 B are for illustrating that the image processing apparatus according to the embodiment of the present disclosure is carried out
Example process explanation figure.Fig. 2A is shown will be by filling according to the image procossing of the embodiment of the present disclosure
An example of the contour images of processing is put, and Fig. 2 B are schematically shown by real according to the disclosure
Apply example image processing apparatus the contour images shown in Fig. 2A are handled obtained by example segmentations
As a result.
As shown in Figure 2 B, can be by Fig. 2A using the image processing apparatus according to the embodiment of the present disclosure
In image contours segmentation for the difference composition on profile contour segment, wherein different contour segments
Represented with different gray scales.It note that for the sake of understanding for diagram, to figure in Fig. 2 B example results
Point on the profile of image in 2A is exaggerated and replicated.
Compared to the communication line obtained in the prior art from the profile of image, the image of the embodiment of the present disclosure
The contour segment that processing unit is obtained provides the segmentation result of the shape based on profile, i.e. obtain
The more accurate contour feature of image, so as to be conducive to object identification, Stereo matching, text analyzing etc.
Various subsequent treatments.
In a preferred embodiment, the first cluster cell 101 of image processing apparatus 10 can be
The angle of candidate's straight line and the angle of the selected straight line are represented in polar coordinate system.Now, in pin
The average value of the angle of candidate's straight line in the cluster obtained to a predetermined point on the profile
The angle of candidate's straight line in another cluster obtained close to 90 degree and for the predetermined point it is flat
Average close to -90 degree in the case of, the first cluster cell 101 by one cluster each time
The angle of straight line is selected to be individually subtracted after 180 degree or each candidate in another described cluster is straight
The angle of line is added after 180 degree respectively, and one cluster and another described cluster are merged into one
New cluster, and the average value of the angle of candidate's straight line in the new cluster is calculated, using as the predetermined point
Selected straight line angle.
In polar coordinate system, it is possible to use on minimum distance and straight line of the straight line away from origin away from origin most
Between near point and the line and pole axis (pole axis corresponds to the X-axis in XY coordinate systems) of origin
Angle is represented to boning out, wherein, depending on the setting of polar coordinate system, above-mentioned line and pole axis it
Between the span of angle can be -90 degree between 90 degree, or can be 0 degree to 180
Between.
Value model away from the angle between the nearest point of origin and the line and pole axis of origin on straight line
- 90 degree are trapped among in the case of between 90 degree, due to the characteristic of polar coordinate system, are caused by set point
, the angle in polar coordinate system be close to respectively 90 degree and -90 degree two straight lines although actually
May closely, but its differential seat angle in polar coordinates is very big.In the span of above-mentioned angle
At 0 degree in the case of between 180 degree, then pass through set point, angle point in polar coordinate system
Although two straight lines for not being close to 0 degree and 180 degree actually may closely, it is in pole
Differential seat angle in coordinate is very big.It will be spent below with the span of above-mentioned angle -90 between 90 degree
In case of illustrate, it will be appreciated by a person skilled in the art that associated description is in above-mentioned angle
Span at 0 degree to being similarly applicable in the case of between 180 degree, thus eliminate repetition and say
It is bright.
Fig. 3 is shown on straight line away from the angle between the nearest point of origin and the line and pole axis of origin
Span in -90 degree a to example in the case of between 90 degree.As shown in figure 3, logical
Cross among set point P straight line L1, L2, L3, straight line L2, L3 it is upper away from the nearest point of origin with
The line of origin and the angle of X-axis are respectively close to 90 degree and -90 degree, i.e. straight line L2, L3 is in pole
Angle in coordinate is spent close to 90 degree and -90 respectively, thus this angle of two straight lines in polar coordinates
Difference it is very big, but straight line L2, L3 actually very close to.
In this case, the above-mentioned processing carried out using the first cluster cell 101, will can pass through
Set point, the angle in polar coordinates be close to respectively 90 degree and -90 degree two candidate's straight lines
(such as straight line L2, L3 in Fig. 3) is placed in a cluster, to avoid because of polar coordinate system in itself
Characteristic and cause candidate's straight line each other actually closely being divided into different clusters.
In one example, the first cluster cell 101 can by the angle in polar coordinates with 90 degree (or
- 90 degree) difference be less than the straight line of predetermined threshold and be considered as angle close to the straight line of 90 degree (or -90 spend),
Or the angle in polar coordinates and the ratio of 90 degree (or -90 degree) can be less than the straight of predetermined threshold
Line is considered as angle close to the straight line of 90 degree (or -90 degree).Those skilled in the art will be seen that, can
The angle of the straight line in polar coordinates is set close to 90 degree depending on application needs and various design factors
The standard and/or dependent thresholds of (or -90 degree), are not described in detail herein.
In a preferred embodiment, the second cluster cell 102 of image processing apparatus 10 can also
The angle of the selected straight line is represented in polar coordinate system.Now, in 102 pairs of wheels of the second cluster cell
After predetermined point on exterior feature is clustered, the angle on the profile, selected straight line is approached
90 degree of first kind point, the second cluster cell 102 can be by the described selected straight of the first kind point
The angle of line subtracts 180 degree, and described selected after 180 degree according to being subtracted for the first kind point
The position of the angle of straight line and first kind point, first kind point cluster is single to the second cluster
In the cluster that member 102 has been obtained.Or, for the angle close -90 of on the profile, selected straight line
The Equations of The Second Kind point of degree, the second cluster cell 102 can be by the selected straight line of the Equations of The Second Kind point
Angle adds 180 degree, and according to the selected straight line being coupled with after 180 degree of the Equations of The Second Kind point
Angle and Equations of The Second Kind point position, by Equations of The Second Kind point cluster to the second cluster cell
In 102 obtained clusters.
As reference picture 3 is discussed before this, in polar coordinate system, due to the characteristic of coordinate system in itself,
Cause by set point, angle respectively close to 90 degree and -90 two straight lines spent although its differential seat angle
It is very big, but actually may be closely.Therefore, the above-mentioned of the second cluster cell 102 progress is utilized
Processing, will can each other be closely located to and select the predetermined point that straight line belongs to above-mentioned situation on profile
It is placed in a cluster, to avoid because the characteristic of polar coordinate system in itself causes position each other and its selected
The point of the angle of straight line all closely is divided into different clusters.
Another example of image processing apparatus according to the embodiment of the present disclosure is described referring to Fig. 4
Structure.Fig. 4 is another example for schematically showing the image processing apparatus according to the embodiment of the present disclosure
The block diagram of structure.
As shown in Figure 4, in image processing apparatus 40, except gathering respectively with first in Fig. 1
Class unit 101 first cluster cell 401 and second cluster cell similar with the second cluster cell 102
Outside 402, in addition to pretreatment unit 400, the pretreatment unit 400 can be configured as in institute
Before the first cluster cell 401 is stated for each predetermined point progress cluster on the profile, sentence
Whether the difference between the maximum angle and minimum angles of candidate's straight line of the open close mistake point is less than predetermined threshold
Value, and when the difference is less than the predetermined threshold, will be carried out by the angle of candidate's straight line of the point
It is average, using the angle as the selected straight line for the point.
Fig. 5 A show that the difference between the maximum angle and minimum angles of candidate's straight line is less than predetermined threshold
An example.As shown in Figure 5A, in candidate's straight line L1, L2, L3 by set point P,
Differential seat angle between each bar straight line is smaller, i.e. pretreatment unit 400 judges that angle is minimum and maximum
Differential seat angle between straight line L1, L3 is less than predetermined threshold, and therefore, pretreatment unit 400 can
So that directly candidate's straight line L1, L2, L3 angle to be averaged, using as the selected of point P
The angle of straight line.
On the other hand, if pretreatment unit 400 judges candidate's straight line by the predetermined point on profile
Maximum angle and minimum angles between difference be more than or equal to predetermined threshold, then can by first cluster
Unit 401 in the way of having described before this by by candidate's straight line of the predetermined point according to these candidates
The angle of straight line is clustered, and then obtains based on cluster result the selected straight line of the point.
It is pre- that Fig. 5 B show that the difference between the maximum angle and minimum angles of candidate's straight line is more than or equal to
Determine an example of threshold value.As shown in Figure 5 B, by set point P candidate's straight line L1, L2,
In L3, the differential seat angle between each bar straight line is larger, i.e. pretreatment unit 400 judges that angle is maximum
Differential seat angle between minimum straight line L1, L3 is more than or equal to predetermined threshold, therefore, it can by
First cluster cell 401 in the way of having described by candidate's straight line L1, L2, L3 according to its angle before this
Degree is clustered, and then obtains the selected straight line for point P based on cluster result.
In an alternative example, the maximum except judging candidate's straight line by the predetermined point on profile
Between angle and minimum angles difference whether be more than or equal to the first predetermined threshold outside (or substitution should
Judge), pretreatment unit 400 can also judge by any one candidate's straight line of the predetermined point with
Whether the second predetermined threshold (the is more than or equal to by the differential seat angle of other candidate's straight lines of the predetermined point
Two predetermined thresholds are less than the first predetermined threshold).
When judge by candidate's straight line of predetermined point with by the predetermined point one or more its
When the differential seat angle of his candidate's straight line is less than the second predetermined threshold, pretreatment unit 400 can wait these
Select the angle of straight line to be averaged, and utilize straight line with the average angle, by the predetermined point
Replace these candidate's straight lines of script as new candidate's straight line, to be supplied to the first cluster cell 401
Handled.For the angle by other any candidate's straight lines predetermined point and that pass through the predetermined point
Degree difference is both greater than or equal to candidate's straight line of the second predetermined threshold, and pretreatment unit 400 can be pressed
It is supplied to the first cluster cell 401 to be handled as former state.
Next, pretreatment unit 400 can be provided by the first cluster cell 401, by wheel
Predetermined point on exterior feature candidate's straight line (including by above-mentioned averaging handle obtain, for replacing original
New candidate's straight line of this candidate's straight line) entered according to the angle of these straight lines in the way of having described before this
Row cluster, and then obtain based on cluster result the selected straight line of the point.
In the above-mentioned processing of pretreatment unit 400, those skilled in the art, which can depend on application, to be needed
Will and various design factors come select correlation predetermined threshold, be not described in detail herein.
Using the image processing apparatus 40 with above-mentioned pretreatment unit 400, profile can passed through
On predetermined point candidate straight line maximum angle and minimum angles between difference be less than predetermined threshold
In the case of, the clustering processing to candidate's straight line is omitted, simply by the angle progress to candidate's straight line
The average selected straight line to obtain the point.
The another example of image processing apparatus according to the embodiment of the present disclosure is described referring to Fig. 6
Structure.Fig. 6 is the another example for schematically showing the image processing apparatus according to the embodiment of the present disclosure
The block diagram of structure.
As shown in Figure 6, in image processing apparatus 60, except gathering respectively with first in Fig. 1
The cluster cell 102 of class unit 101 and second similar the first cluster cell 601 and the second cluster cell
Outside 602, in addition to acquiring unit 600, the acquiring unit 600 is for being based in pending image
Profile obtain candidate's straight line by the predetermined point on the profile.
In a preferred embodiment, can set may be by pending image for acquiring unit 600
Profile on predetermined point a plurality of potential straight line, and build corresponding accumulator matrix.It is cumulative at this
In device matrix, each element corresponds to a potential straight line, and the element value of the element is potential corresponding to this
The number for the predetermined point on profile that straight line is passed through.Acquiring unit 600 can be by accumulator matrix
, element value be more than accumulation threshold those elements corresponding to potential straight line be used as pass through pending figure
Candidate's straight line of the predetermined point on profile as in, so that the first cluster cell 601 is handled.
More specifically, in the preferred embodiment, acquiring unit 600 can be obtained by following manner
Pass through candidate's straight line of the predetermined point on the profile:
MXn accumulator matrix is built, in the accumulator matrix, the index i per a line is
The predetermined quantization parameter for representing distance, the index j of each row is predetermined expression angle
Quantization parameter, may pass through one of the predetermined point on each element (i, j) the correspondence profile
Potential straight line and with unified finite element value (being, for example, 0), the potential straight line is away from origin
Minimum distance is the point and origin away from origin recently on the distance represented by index i, the potential straight line
Angle between line and X-axis is the angle represented by index j, wherein, i, j, m, n are certainly
So number, and 0<I≤m, 0<j≤n;
For each predetermined point on the profile, for the index of each row of the accumulator matrix
Angle, θ corresponding to jj, calculate and angle, θjMost low coverage of the corresponding straight line by the point away from origin
From ρj, and by the distance corresponding in the element of the row, line index i and the most low coverage that is calculated
From ρjThe element value of that immediate element is incremented by one;And
Element value in the accumulator matrix is more than potential straight corresponding to the element of accumulation threshold
Line is used as candidate's straight line by the predetermined point on the profile.
In the above-mentioned processing of acquiring unit 600, with minimum distance of the straight line away from origin and straight line
Boning out is represented to away from the angle between the nearest point of origin and the line and X-axis of origin.So,
Each element (i, j) in accumulator matrix can correspond to following straight lines:The straight line is away from origin
Minimum distance is the distance represented by quantization parameter i, and the angle between the straight line and X-axis is amount
Change the angle represented by coefficient j.
For a predetermined point P on the profile in pending image, acquiring unit 600 be directed to
Surely the corresponding angle, θs of quantization parameter j arrangedjTo determine and the angle, θjIt is corresponding to pass through the straight of point P
The minimum distance ρ of line-spacing originj, i.e. determine by minimum distance ρjAnd angle, θjWhat is limited passes through
The straight line of the point.Then, in the potential straight line corresponding to the element in the given row of accumulator, obtain
Unit 600 is taken to determine and by above-mentioned minimum distance ρjAnd angle, θjThe straight line of restriction is immediate potential
Straight line, i.e. determine accumulator the given row element in, distance corresponding to line index i with
The minimum distance ρ calculatedjThat immediate element (i, j).Acquiring unit 600 will be determined
Element (i, j) corresponding to potential straight line be considered as may pass through predetermined point P straight line so that will
The element value of the element (i, j) in accumulator adds 1.Institute of the acquiring unit 600 for accumulator
There are row to perform the above-mentioned processing for predetermined point P.
In this way, acquiring unit 600 is for each predetermined click-through on the profile in pending image
Row processing, so as to obtain final accumulator matrix.
In final accumulator matrix, element value can be more than accumulation threshold by acquiring unit 600
Potential straight line corresponding to element waits these as candidate's straight line by the predetermined point on profile
Straight line is selected to be supplied to the first cluster cell 601 to be handled.Those skilled in the art will be seen that, can
It is no longer detailed herein suitably to set accumulation threshold based on application demand and other factors etc. that are related to
Description.
In a further advantageous embodiment, acquiring unit 600 can be obtained by following manner passes through institute
State candidate's straight line of the predetermined point on profile, i.e. acquiring unit 600 is for each on the profile
Predetermined point:Build from the predetermined point, other predetermined points connecting respectively on the profile
A plurality of potential straight line;For the every potential straight line built for the predetermined point, if on the profile,
Distance of at least two predetermined points away from the potential straight line that mutual distance is less than the first distance threshold is small
In second distance threshold value, then the potential straight line is used as to candidate's straight line by the predetermined point.
Connect from this point, respectively for what each predetermined point on profile was built in acquiring unit 600
After a plurality of potential straight line of an other predetermined points on cock wheel exterior feature, it is understood that there may be following situations:Wheel
Two predetermined points located far away from one another may accidentally fall on exterior feature (for example exists on the potential straight line of same
In sine curve y=sin (x), x is respectively that 0 and 180 degree, y values all fall within X for 0 two points
On axle), but the potential straight line does not reflect the feature of profile well.
In the preferred embodiment, define acquiring unit 600 and candidate's straight line is determined from potential straight line
Condition be:On profile, mutual distance is less than at least two predetermined points of the first distance threshold away from this
The distance of potential straight line is respectively less than second distance threshold value.Such condition is conducive to excluding above-mentioned, simultaneously
The potential straight line of the feature of profile is not reflected well.
In addition, those skilled in the art will be seen that, acquiring unit 600 can be (all using known technology
Such as straight-line detection technology) to obtain by predetermined on the profile based on the profile in pending image
Candidate's straight line of point, is no longer described in detail here.
The image processing apparatus and its group according to the embodiment of the present disclosure are described above by reference to Fig. 1 to Fig. 6
Into unit and relevant treatment.Using the image processing apparatus according to the embodiment of the present disclosure, it will can scheme
The contours segmentation of arbitrary shape as in is into contour segment, in favor of the wheel by the use of segmentation result as image
Wide feature carries out subsequent treatment.
According to another aspect of the disclosure, there is provided a kind of image processing method.Fig. 7 is schematic
Ground shows the flow chart of the example flow of the image processing method according to the embodiment of the present disclosure.
It can be used for according to the image processing method of the embodiment of the present disclosure based on the wheel in pending image
Exterior feature extracts contour segment.As shown in fig. 7, image processing method 70 can include:First sorting procedure
S701, for for each predetermined point on the profile, by it is being obtained based on the profile, pass through
Candidate's straight line of the point is one or more clusters according to the angle of candidate's straight line cluster, and is counted respectively
The average value of the angle of candidate's straight line in each cluster in one or more of clusters is calculated, to make respectively
For the angle of one or more selected straight lines for the point;And the second sorting procedure S703, use
In by the predetermined point on the profile according to the angle of the selected straight line of each predetermined point and the point
Position is clustered, and by the point in resulting each cluster using in units of cluster as from the profile
The contour segment of extraction, using the contour feature as the pending image.
In a preferred embodiment, each predetermined point on for the profile will pass through the point
During candidate's straight line cluster is for cluster and the step of calculate the angle for the selected straight line of the point, Ke Yi
The angle of candidate's straight line and the angle of the selected straight line are represented in polar coordinate system, wherein, in pin
The average value of the angle of candidate's straight line in the cluster obtained to a predetermined point on the profile
The angle of candidate's straight line in another cluster obtained close to 90 degree and for the predetermined point it is flat
In the case that average is spent close to -90, distinguish by the angle of each candidate's straight line in one cluster
Subtract after 180 degree or added respectively by the angle of each candidate's straight line in another described cluster
After 180 degree, one cluster and another described cluster are merged into a new cluster, and it is new to calculate this
The average value of the angle of candidate's straight line in cluster, using the angle as the selected straight line for the predetermined point
Degree.
In a preferred embodiment, in the step of being clustered the predetermined point on the profile,
The angle of the selected straight line, also, image processing method 70 can also be represented in polar coordinate system
Can also it include:After the step of being clustered the predetermined point on the profile, for the wheel
The angle of on exterior feature, selected straight line is close to 90 degree of first kind point, by described in the first kind point
The angle of selected straight line subtracts 180 degree, and according to the institute being subtracted after 180 degree of the first kind point
The angle of selected straight line and the position of first kind point are stated, by first kind point cluster to having led to
Cross in cluster obtained from the predetermined point on the profile is clustered;Or for it is on the profile,
The angle of selected straight line is close to -90 Equations of The Second Kind points spent, by the selected straight line of the Equations of The Second Kind point
Angle add 180 degree, it is and described selected straight after 180 degree according to being coupled with for the Equations of The Second Kind point
The position of the angle of line and Equations of The Second Kind point, by Equations of The Second Kind point cluster to by by described in
In cluster obtained from predetermined point on profile is clustered.
Above-mentioned image processing method 70 and its each step can be realized by above by reference to Fig. 1 to Fig. 3
The processing that the image processing apparatus 10 of description and its each component units are carried out, and realize similar effect
Really, repeat specification is no longer carried out herein.
Another example of image processing method according to the embodiment of the present disclosure is described referring to Fig. 8
Flow.Fig. 8 is another example for schematically showing the image processing method according to the embodiment of the present disclosure
The flow chart of flow.
As shown in figure 8, in image processing method 80, except being clustered respectively with first in Fig. 7
The cluster steps of the first similar with the second sorting procedure S703 step S701 sorting procedure S801 and second
Outside rapid S803, in addition to pre-treatment step S800.That is, in image processing method 80,
For each predetermined point on the profile by the step of being cluster by candidate's straight line of point cluster it
Before, carry out following pre-treatment step S800:Judge by the maximum angle of candidate's straight line of the point and
Whether the difference between minimum angles is less than predetermined threshold, and when the difference is less than the predetermined threshold,
The angle of candidate's straight line by the point is averaged, using the angle as the selected straight line for the point
Degree.
Above-mentioned image processing method 80 and its each step can be realized by above by reference to Fig. 4 to Fig. 5 B
The processing that the image processing apparatus 40 of description and its each component units are carried out, and realize similar effect
Really, repeat specification is no longer carried out herein.
The another example of image processing method according to the embodiment of the present disclosure is described referring to Fig. 9
Flow.Fig. 9 is the another example for schematically showing the image processing method according to the embodiment of the present disclosure
The flow chart of flow.
As shown in figure 9, in example image processing method 90, except respectively with first in Fig. 7
The first sorting procedure S901 and second similar with the second sorting procedure S703 sorting procedure S701 gathers
Outside class step S903, in addition to obtaining step S900.That is, in example image processing method 90
In, each predetermined point on for the profile is by step of the candidate's straight line cluster by the point for cluster
Before rapid, following obtaining step S900 are performed:Obtained based on the profile by the profile
Candidate's straight line of predetermined point.
In a preferred embodiment, in obtaining step S900, obtained and passed through by following manner
Candidate's straight line of predetermined point on the profile:
MXn accumulator matrix is built, in the accumulator matrix, the index i per a line is
The predetermined quantization parameter for representing distance, the index j of each row is predetermined expression angle
Quantization parameter, may pass through one of the predetermined point on each element (i, j) the correspondence profile
Potential straight line and with unified finite element value (being, for example, 0), the potential straight line is away from origin
Minimum distance is the point and origin away from origin recently on the distance represented by index i, the potential straight line
Angle between line and X-axis is the angle represented by index j, wherein, i, j, m, n are certainly
So number, and 0<I≤m, 0<j≤n;
For each predetermined point on the profile, for the index of each row of the accumulator matrix
Angle, θ corresponding to jj, calculate and angle, θjMost low coverage of the corresponding straight line by the point away from origin
From ρj, and by the distance corresponding in the element of the row, line index i and the most low coverage that is calculated
From ρjThe element value of that immediate element is incremented by one;And
Element value in the accumulator matrix is more than potential straight corresponding to the element of accumulation threshold
Line is used as candidate's straight line by the predetermined point on the profile.
In a further advantageous embodiment, in obtaining step S900, obtain logical by following manner
The candidate's straight line for the predetermined point crossed on the profile, i.e. for each predetermined point on the profile:
Build a plurality of potential straight line of the other points connected from this point, respectively on the profile;Pin
To the every potential straight line built for the point, if on the profile, mutual distance be less than first away from
From 1 points of threshold value away from second distance threshold value is respectively less than with a distance from the potential straight line, then this is dived
Candidate's straight line by the point is used as in straight line.
Above-mentioned image processing method 90 and its each step can be realized by being described above by reference to Fig. 6
The processing that image processing apparatus 60 and its each component units are carried out, and similar effect is realized,
This no longer carries out repeat specification.
According to the another aspect of the disclosure, there is provided a kind of image processing equipment.Figure 10 is signal
The block diagram of the exemplary construction of image processing equipment according to the embodiment of the present disclosure is shown to property.
It can be used for according to the image processing equipment of the embodiment of the present disclosure based on the wheel in pending image
Exterior feature extracts contour segment.As shown in Figure 10, image processing equipment 1000 can include controller 1001.
Controller 1001 can be configured as:, will be based on described for each predetermined point on the profile
It is that profile is obtained, by candidate's straight line of the point according to the angle of candidate's straight line cluster for one or
Multiple clusters, and calculate the angle of candidate's straight line in each cluster in one or more of clusters respectively
Average value, using the angle as one or more selected straight lines for the point;And will be described
Predetermined point on profile is carried out according to the angle of the selected straight line of each predetermined point and the position of the point
Cluster, and it regard the point in resulting each cluster as the wheel extracted from the profile using in units of cluster
Wide section, using the contour feature as the pending image.
Image processing equipment 1000 can utilize arbitrary specialized hardware, special-purpose computer or general
General purpose personal computer realizes, and controller 1001 can using CPU (CPU),
The various appropriate devices such as processor, application specific integrated circuit are realized.
Utilize image processing equipment 1000, it is possible to achieve by the figure described above by reference to Fig. 1 to Fig. 6
As the processing that processing unit 10,40,60 and its each component units are carried out, and obtain corresponding
Effect, is lack of repetition herein.
Figure 11 is to show to can be used to realize image processing apparatus according to the embodiment of the present disclosure, method
And a kind of structure diagram of possible hardware configuration of equipment.
In fig. 11, CPU (CPU) 1101 is according in read-only storage (ROM) 1102
The program of storage or the program that random access memory (RAM) 1103 is loaded into from storage part 1108
Perform various processing.In RAM 1103, always according to need store when CPU 1101 perform it is various
Required data during processing etc..CPU 1101, ROM 1102 and RAM 1103 are via bus
1104 are connected to each other.Input/output interface 1105 is also connected to bus 1104.
Components described below is also connected to input/output interface 1105:Importation 1106 (including keyboard,
Mouse etc.), output par, c 1107 (including display, such as cathode-ray tube (CRT), liquid crystal
Display (LCD) etc., and loudspeaker etc.), storage part 1108 (including hard disk etc.), communications portion
1109 (including NIC is such as LAN card, modems).Communications portion 1109 is passed through
Communication process is performed by network such as internet.As needed, driver 1110 can be connected to defeated
Enter/output interface 1105.Detachable media 1111 such as disk, CD, magneto-optic disk, semiconductor are deposited
Reservoir etc. can be installed on driver 1110 as needed so that the computer read out
Program can be installed in storage part 1108 as needed.
In addition, the disclosure also proposed a kind of program production of instruction code for the machine-readable that is stored with
Product.When above-mentioned instruction code is read and performed by machine, it can perform above-mentioned according to the embodiment of the present disclosure
Image processing method.Correspondingly, for carrying such as disk, CD, the magneto-optic of this program product
The various storage mediums of disk, semiconductor memory etc. are also included within the disclosure of the disclosure.
Above in the description of disclosure specific embodiment, for a kind of description of embodiment and/or
The feature shown can be made in same or similar mode in one or more other embodiments
With, it is combined with feature in other embodiment, or substitute the feature in other embodiment.
In addition, the method for the presently disclosed embodiments be not limited to specifications described in or accompanying drawing
In the time sequencing that shows perform, can also according to other time sequencings, concurrently or independently
Perform.Therefore, the execution sequence of the method described in this specification technical scope structure not of this disclosure
Into limitation.
It should be further understood that can also be to be stored according to each operating process of the above method of the disclosure
The mode of computer executable program in various machine readable storage mediums is realized.
Moreover, the purpose of the disclosure can also be accomplished in the following manner:It will be stored with above-mentioned executable
The storage medium of program code is directly or indirectly supplied to system or equipment, and the system or sets
Computer or CPU (CPU) in standby read and perform said procedure code.
Now, as long as the system or equipment have the function of configuration processor, then the embodiment party of the disclosure
Formula is not limited to program, and the program can also be arbitrary form, for example, target program, solution
Release the program of device execution or be supplied to shell script of operating system etc..
These above-mentioned machinable mediums include but is not limited to:Various memories and memory cell,
Semiconductor equipment, disk cell such as light, magnetic and magneto-optic disk, and other Jie suitable for storage information
Matter etc..
In addition, customer information processing terminal is by the corresponding website that is connected on internet, and will be according to
Downloaded according to the computer program code of the disclosure and be installed in the information processing terminal and then perform the journey
Sequence, can also realize the presently disclosed embodiments.
To sum up, according in the embodiment of the present disclosure, present disclose provides following scheme, but not limited to this:
A kind of image processing apparatus of scheme 1., for based on the contours extract profile in pending image
Section, described device includes:
First cluster cell, it will be obtained for each predetermined point on the profile based on the profile
, by candidate's straight line of the point according to the angle of candidate's straight line cluster be one or more
Cluster, and calculate the angle of candidate's straight line in each cluster in one or more of clusters respectively and be averaged
Value, using the angle as one or more selected straight lines for the point;And
Second cluster cell, it is by selected straight line of the predetermined point on the profile according to each predetermined point
Angle and the position of the point clustered, and by the point in resulting each cluster in units of cluster
As the contour segment extracted from the profile, using the contour feature as the pending image.
Image processing apparatus of the scheme 2. as described in scheme 1, wherein,
First cluster cell represents the angle of candidate's straight line and the choosing in polar coordinate system
The angle of boning out,
Wherein, candidate's straight line in the cluster that a predetermined point on for the profile is obtained
Candidate in another cluster that the average value of angle is obtained close to 90 degree and for the predetermined point is straight
The average value of the angle of line is close in the case of -90 degree, and first cluster cell is will be one
The angle of each candidate's straight line in cluster is individually subtracted after 180 degree or by another described cluster
Each candidate's straight line angle respectively add 180 degree after, by one cluster with it is described another
Cluster merges into a new cluster, and calculates the average value of the angle of candidate's straight line in the new cluster, using as
For the angle of the selected straight line of the predetermined point.
Image processing apparatus of the scheme 3. as described in scheme 1 or 2, wherein,
Second cluster cell represents the angle of the selected straight line in polar coordinate system, and
After second cluster cell carries out the cluster,
For on the profile, selected straight line angle close to 90 degree of first kind point, described the
The angle of the selected straight line of the first kind point is subtracted 180 degree by two cluster cells, and according to institute
State the angle and first kind point of the selected straight line being subtracted after 180 degree of first kind point
Position, the first kind point is clustered in the cluster that has been obtained to second cluster cell;Or
For on the profile, selected straight line angle close to -90 degree Equations of The Second Kind points, described the
The angle of the selected straight line of the Equations of The Second Kind point is added 180 degree by two cluster cells, and according to institute
State the angle and Equations of The Second Kind point of the selected straight line being coupled with after 180 degree of Equations of The Second Kind point
Position, the Equations of The Second Kind point is clustered in the cluster that has been obtained to second cluster cell.
Image processing apparatus of the scheme 4. as described in scheme 1, in addition to:
Pretreatment unit, the pretreatment unit is configured as in first cluster cell for described
Each predetermined point on profile is carried out before the cluster, judges the maximum of candidate's straight line by the point
Whether the difference between angle and minimum angles is less than predetermined threshold, and is less than the predetermined threshold in the difference
During value, the angle of candidate's straight line by the point is averaged, using as the selected straight of the point
The angle of line.
Image processing apparatus of the scheme 5. as described in scheme 1, in addition to:
Acquiring unit, it obtains straight by the candidate of the predetermined point on the profile based on the profile
Line.
Image processing apparatus of the scheme 6. as described in scheme 5, wherein, the acquiring unit is under
The mode of stating obtains candidate's straight line by the predetermined point on the profile:
Build the predetermined point that each element in accumulator matrix, the matrix corresponds on the profile
The potential straight line that may pass through, the line index of the element corresponds to the potential straight line away from origin most
Closely, the column index of the element corresponds to the company away from the nearest point of origin and origin on the potential straight line
Each element in angle between line and X-axis, the matrix has unified finite element value;
For each predetermined point on the profile, for each row of the accumulator matrix, it is determined that
Work as by the point, straight line with the angle corresponding to the column index of the row, and by the element of the row
In, the element value of corresponding potential straight line and that immediate element of identified straight line is incremented by
One;And
Element value in the accumulator matrix is more than potential straight corresponding to the element of accumulation threshold
Line is used as candidate's straight line by the predetermined point on the profile.
Image processing apparatus of the scheme 7. as described in scheme 5, wherein, the acquiring unit is under
The mode of stating obtains candidate's straight line by the predetermined point on the profile:
For each predetermined point on the profile,
Structure connects a plurality of of an other predetermined points on the profile from this point, respectively
Potential straight line;
For the every potential straight line built for the point, if on the profile, mutual distance
Distance of at least two predetermined points away from the potential straight line less than the first distance threshold is respectively less than
Two distance thresholds, then regard the potential straight line as candidate's straight line by the point.
A kind of image processing method of scheme 8., for based on the contours extract profile in pending image
Section, methods described includes:
, will be being obtained based on the profile, by the point for each predetermined point on the profile
Candidate's straight line is one or more clusters according to the angle of candidate's straight line cluster, and is calculated respectively described
The average value of the angle of candidate's straight line in each cluster in one or more clusters, using as
The angle of one or more selected straight lines of the point;And
By the predetermined point on the profile according to the angle of the selected straight line of each predetermined point and the point
Position clustered, and by the point in resulting each cluster using in units of cluster as from the profile
The contour segment of middle extraction, using the contour feature as the pending image.
Image processing method of the scheme 9. as described in scheme 8, wherein,
Each predetermined point on for the profile will be clustered as cluster simultaneously by candidate's straight line of the point
In the step of calculating the angle for the selected straight line of the point:
The angle of candidate's straight line and the angle of the selected straight line are represented in polar coordinate system,
Wherein, candidate's straight line in the cluster that a predetermined point on for the profile is obtained
Candidate in another cluster that the average value of angle is obtained close to 90 degree and for the predetermined point is straight
In the case that the average value of the angle of line is spent close to -90, each candidate in one cluster is straight
The angle of line is individually subtracted after 180 degree or by each candidate's straight line in another described cluster
Angle is added after 180 degree respectively, and one cluster and another described cluster are merged into one newly
Cluster, and the average value of the angle of candidate's straight line in the new cluster is calculated, using as the predetermined point
The angle of selected straight line.
Image processing method of the scheme 10. as described in scheme 8 or 9, wherein,
In the step of being clustered the predetermined point on the profile, represent described in polar coordinate system
The angle of selected straight line, and
Methods described also includes:After the step of being clustered the predetermined point on the profile,
For angle on the profile, selecting straight line close to 90 degree of first kind point, incite somebody to action described
The angle of the selected straight line of first kind point subtracts 180 degree, and being subtracted according to the first kind point
The angle of the selected straight line gone after 180 degree and the position of first kind point, by described first
Class point cluster is into cluster obtained from by the way that the predetermined point on the profile is clustered;Or
, will be described for the Equations of The Second Kind points of close -90 degree of angle of on the profile, selected straight line
The angle of the selected straight line of Equations of The Second Kind point adds 180 degree, and being added according to the Equations of The Second Kind point
The angle of the selected straight line after upper 180 degree and the position of Equations of The Second Kind point, by described second
Class point cluster is into cluster obtained from by the way that the predetermined point on the profile is clustered.
Image processing method of the scheme 11. as described in scheme 8, in addition to:
Each predetermined point on for the profile will be clustered as cluster by candidate's straight line of the point
Before step, following pretreatments are carried out:The maximum angle and minimum for candidate's straight line that judgement passes through the point
Whether the difference between angle is less than predetermined threshold, and when the difference is less than the predetermined threshold, will be logical
The angle for crossing candidate's straight line of the point is averaged, using the angle as the selected straight line for the point.
Image processing method of the scheme 12. as described in scheme 8, in addition to:
Each predetermined point on for the profile will be clustered as cluster by candidate's straight line of the point
Before step, candidate's straight line by the predetermined point on the profile is obtained based on the profile.
Image processing method of the scheme 13. as described in scheme 12, wherein, obtained by following manner
Pass through candidate's straight line of the predetermined point on the profile:
Build the predetermined point that each element in accumulator matrix, the matrix corresponds on the profile
The potential straight line that may pass through, the line index of the element corresponds to the potential straight line away from origin most
Closely, the column index of the element corresponds to the company away from the nearest point of origin and origin on the potential straight line
Each element in angle between line and X-axis, the matrix has unified finite element value;
For each predetermined point on the profile, for each row of the accumulator matrix, it is determined that
Work as by the point, straight line with the angle corresponding to the column index of the row, and by the element of the row
In, the element value of corresponding potential straight line and that immediate element of identified straight line is incremented by
One;And
Element value in the accumulator matrix is more than potential straight corresponding to the element of accumulation threshold
Line is used as candidate's straight line by the predetermined point on the profile.
Image processing method of the scheme 14. as described in scheme 12, wherein, obtained by following manner
Pass through candidate's straight line of the predetermined point on the profile:
For each predetermined point on the profile,
Structure connects a plurality of potential of an other points on the profile from this point, respectively
Straight line;
For the every potential straight line built for the point, if on the profile, mutual distance
Less than the first distance threshold 1 points of distances away from the potential straight line be respectively less than second away from
From threshold value, then the potential straight line is used as to candidate's straight line by the point.
A kind of image processing equipment of scheme 15., for based on the contours extract profile in pending image
Section, the equipment includes:
Controller, the controller is configured as:
For each predetermined point on the profile, by it is being obtained based on the profile, pass through this
Candidate's straight line of point is one or more clusters according to the angle of candidate's straight line cluster, and respectively
The average value of the angle of candidate's straight line in each cluster in one or more of clusters is calculated, with
Respectively as the angle of one or more selected straight lines for the point;And
By the predetermined point on the profile according to the angle of the selected straight line of each predetermined point and
The position of the point is clustered, and by the point in resulting each cluster using in units of cluster as from
The contour segment extracted in the profile, using the contour feature as the pending image.
Finally, in addition it is also necessary to explanation, in the disclosure, such as first and second or the like relation
Term is used merely to make a distinction an entity or operation with another entity or operation, and differs
Provisioning request implies between these entities or operation there is any this actual relation or order.
Moreover, term " comprising ", "comprising" or any other variant thereof is intended to cover non-exclusive inclusion,
So that process, method, article or equipment including a series of key elements not only include those key elements,
But also other key elements including being not expressly set out, or also include for this process, method,
Article or the intrinsic key element of equipment.In the absence of more restrictions, by sentence " including one
It is individual ... " limit key element, it is not excluded that the process including the key element, method, article or
Also there is other identical element in equipment.
Although being had been disclosed above by the description of the specific embodiment of the disclosure to the disclosure,
However, it is to be understood that those skilled in the art can design in the spirit and scope of the appended claims
Various modifications of this disclosure, improvement or equivalent.These modifications, improvement or equivalent also should
When being to be considered as included in the claimed scope of the disclosure.
Claims (10)
1. a kind of image processing apparatus, for based on the contours extract contour segment in pending image,
Described device includes:
First cluster cell, it will be obtained for each predetermined point on the profile based on the profile
, by candidate's straight line of the point according to the angle of candidate's straight line cluster be one or more
Cluster, and calculate the angle of candidate's straight line in each cluster in one or more of clusters respectively and be averaged
Value, using the angle as one or more selected straight lines for the point;And
Second cluster cell, it is by selected straight line of the predetermined point on the profile according to each predetermined point
Angle and the position of the point clustered, and by the point in resulting each cluster in units of cluster
As the contour segment extracted from the profile, using the contour feature as the pending image.
2. image processing apparatus as claimed in claim 1, wherein,
First cluster cell represents the angle of candidate's straight line and the choosing in polar coordinate system
The angle of boning out,
Wherein, candidate's straight line in the cluster that a predetermined point on for the profile is obtained
Candidate in another cluster that the average value of angle is obtained close to 90 degree and for the predetermined point is straight
The average value of the angle of line is close in the case of -90 degree, and first cluster cell is will be one
The angle of each candidate's straight line in cluster is individually subtracted after 180 degree or by another described cluster
Each candidate's straight line angle respectively add 180 degree after, by one cluster with it is described another
Cluster merges into a new cluster, and calculates the average value of the angle of candidate's straight line in the new cluster, using as
For the angle of the selected straight line of the predetermined point.
3. image processing apparatus as claimed in claim 1 or 2, wherein,
Second cluster cell represents the angle of the selected straight line in polar coordinate system, and
After second cluster cell carries out the cluster,
For on the profile, selected straight line angle close to 90 degree of first kind point, described the
The angle of the selected straight line of the first kind point is subtracted 180 degree by two cluster cells, and according to institute
State the angle and first kind point of the selected straight line being subtracted after 180 degree of first kind point
Position, the first kind point is clustered in the cluster that has been obtained to second cluster cell;Or
For on the profile, selected straight line angle close to -90 degree Equations of The Second Kind points, described the
The angle of the selected straight line of the Equations of The Second Kind point is added 180 degree by two cluster cells, and according to institute
State the angle and Equations of The Second Kind point of the selected straight line being coupled with after 180 degree of Equations of The Second Kind point
Position, the Equations of The Second Kind point is clustered in the cluster that has been obtained to second cluster cell.
4. image processing apparatus as claimed in claim 1, in addition to:
Pretreatment unit, the pretreatment unit is configured as in first cluster cell for described
Each predetermined point on profile is carried out before the cluster, judges the maximum of candidate's straight line by the point
Whether the difference between angle and minimum angles is less than predetermined threshold, and is less than the predetermined threshold in the difference
During value, the angle of candidate's straight line by the point is averaged, using as the selected straight of the point
The angle of line.
5. image processing apparatus as claimed in claim 1, in addition to:
Acquiring unit, it obtains straight by the candidate of the predetermined point on the profile based on the profile
Line.
6. image processing apparatus as claimed in claim 5, wherein, the acquiring unit is under
The mode of stating obtains candidate's straight line by the predetermined point on the profile:
The each element built in accumulator matrix, the accumulator matrix corresponds on the profile
The potential straight line that predetermined point may pass through, the line index of the element corresponds to the potential straight line away from original
The minimum distance of point, the column index of the element correspond on the potential straight line away from the nearest point of origin with it is former
Each element in angle between the line and X-axis of point, the matrix has unified initial element
Element value;
For each predetermined point on the profile, for each row of the accumulator matrix, it is determined that
By the predetermined point, straight line with the angle corresponding to the column index of the row, and by the member of the row
The element value of among element, corresponding potential straight line and that immediate element of identified straight line
It is incremented by one;And
Element value in the accumulator matrix is more than potential straight corresponding to the element of accumulation threshold
Line is used as candidate's straight line by the predetermined point on the profile.
7. image processing apparatus as claimed in claim 5, wherein, the acquiring unit is under
The mode of stating obtains candidate's straight line by the predetermined point on the profile:
For each predetermined point on the profile,
Structure connects a plurality of of an other predetermined points on the profile from this point, respectively
Potential straight line;
For the every potential straight line built for the point, if on the profile, mutual distance
Distance of at least two predetermined points away from the potential straight line less than the first distance threshold is respectively less than
Two distance thresholds, then regard the potential straight line as candidate's straight line by the point.
8. a kind of image processing method, for based on the contours extract contour segment in pending image,
Methods described includes:
, will be being obtained based on the profile, by the point for each predetermined point on the profile
Candidate's straight line is one or more clusters according to the angle of candidate's straight line cluster, and is calculated respectively described
The average value of the angle of candidate's straight line in each cluster in one or more clusters, using as
The angle of one or more selected straight lines of the point;And
By the predetermined point on the profile according to the angle of the selected straight line of each predetermined point and the point
Position clustered, and by the point in resulting each cluster using in units of cluster as from the profile
The contour segment of middle extraction, using the contour feature as the pending image.
9. image processing method as claimed in claim 8, wherein,
Each predetermined point on for the profile will be clustered as cluster simultaneously by candidate's straight line of the point
In the step of calculating the angle for the selected straight line of the point:
The angle of candidate's straight line and the angle of the selected straight line are represented in polar coordinate system,
Wherein, candidate's straight line in the cluster that a predetermined point on for the profile is obtained
Candidate in another cluster that the average value of angle is obtained close to 90 degree and for the predetermined point is straight
In the case that the average value of the angle of line is spent close to -90, each candidate in one cluster is straight
The angle of line is individually subtracted after 180 degree or by each candidate's straight line in another described cluster
Angle is added after 180 degree respectively, and one cluster and another described cluster are merged into one newly
Cluster, and the average value of the angle of candidate's straight line in the new cluster is calculated, using as the predetermined point
The angle of selected straight line.
10. a kind of image processing equipment, for based on the contours extract contour segment in pending image,
The equipment includes:
Controller, the controller is configured as:
For each predetermined point on the profile, by it is being obtained based on the profile, pass through this
Candidate's straight line of point is one or more clusters according to the angle of candidate's straight line cluster, and respectively
The average value of the angle of candidate's straight line in each cluster in one or more of clusters is calculated, with
Respectively as the angle of one or more selected straight lines for the point;And
By the predetermined point on the profile according to the angle of the selected straight line of each predetermined point and
The position of the point is clustered, and by the point in resulting each cluster using in units of cluster as from
The contour segment extracted in the profile, using the contour feature as the pending image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610197282.8A CN107292893A (en) | 2016-03-31 | 2016-03-31 | Image processing apparatus, image processing method and image processing equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610197282.8A CN107292893A (en) | 2016-03-31 | 2016-03-31 | Image processing apparatus, image processing method and image processing equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107292893A true CN107292893A (en) | 2017-10-24 |
Family
ID=60086869
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610197282.8A Pending CN107292893A (en) | 2016-03-31 | 2016-03-31 | Image processing apparatus, image processing method and image processing equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107292893A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109409176A (en) * | 2018-01-04 | 2019-03-01 | 北京星衡科技有限公司 | A kind of method and apparatus that the plot for remote sensing image is extracted |
CN110705385A (en) * | 2019-09-12 | 2020-01-17 | 深兰科技(上海)有限公司 | Method, device, equipment and medium for detecting angle of obstacle |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003140544A (en) * | 2001-11-07 | 2003-05-16 | Nakanihon Koku Kk | Method for discriminating characteristic land feature and device for the same |
CN104463857A (en) * | 2014-11-27 | 2015-03-25 | 南宁市界围工程咨询有限公司 | Extracting method for structure data in geographic data of oblique aviation photography |
CN104655041A (en) * | 2015-01-05 | 2015-05-27 | 山东理工大学 | Industrial part contour line multi-feature extracting method with additional constraint conditions |
EP2980754A1 (en) * | 2014-07-28 | 2016-02-03 | Thomson Licensing | Method and apparatus for generating temporally consistent superpixels |
-
2016
- 2016-03-31 CN CN201610197282.8A patent/CN107292893A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003140544A (en) * | 2001-11-07 | 2003-05-16 | Nakanihon Koku Kk | Method for discriminating characteristic land feature and device for the same |
EP2980754A1 (en) * | 2014-07-28 | 2016-02-03 | Thomson Licensing | Method and apparatus for generating temporally consistent superpixels |
CN104463857A (en) * | 2014-11-27 | 2015-03-25 | 南宁市界围工程咨询有限公司 | Extracting method for structure data in geographic data of oblique aviation photography |
CN104655041A (en) * | 2015-01-05 | 2015-05-27 | 山东理工大学 | Industrial part contour line multi-feature extracting method with additional constraint conditions |
Non-Patent Citations (1)
Title |
---|
陈月林 等: "基于曲率的轮廓精确分段技术", 《华中理工大学学报》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109409176A (en) * | 2018-01-04 | 2019-03-01 | 北京星衡科技有限公司 | A kind of method and apparatus that the plot for remote sensing image is extracted |
CN110705385A (en) * | 2019-09-12 | 2020-01-17 | 深兰科技(上海)有限公司 | Method, device, equipment and medium for detecting angle of obstacle |
CN110705385B (en) * | 2019-09-12 | 2022-05-03 | 深兰科技(上海)有限公司 | Method, device, equipment and medium for detecting angle of obstacle |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112100387B (en) | Training method and device of neural network system for text classification | |
CN108664897A (en) | Bank slip recognition method, apparatus and storage medium | |
CN109214238A (en) | Multi-object tracking method, device, equipment and storage medium | |
CN111340616B (en) | Method, device, equipment and medium for approving online loan | |
CN107729935B (en) | The recognition methods of similar pictures and device, server, storage medium | |
CN109472626B (en) | Intelligent financial risk control method and system for mobile phone leasing service | |
CN111222976A (en) | Risk prediction method and device based on network diagram data of two parties and electronic equipment | |
CN110019163A (en) | Method, system, equipment and the storage medium of prediction, the recommendation of characteristics of objects | |
CN112562069A (en) | Three-dimensional model construction method, device, equipment and storage medium | |
CN111340506A (en) | Method and device for identifying risk of transaction behavior, storage medium and computer equipment | |
CN110009097A (en) | The image classification method of capsule residual error neural network, capsule residual error neural network | |
CN112560754A (en) | Bill information acquisition method, device, equipment and storage medium | |
CN113780098A (en) | Character recognition method, character recognition device, electronic equipment and storage medium | |
CN113393371A (en) | Image processing method and device and electronic equipment | |
CN107292893A (en) | Image processing apparatus, image processing method and image processing equipment | |
CN109242671A (en) | A kind of credit violation correction method and system based on multi-angle of view deficiency of data | |
CN107767155A (en) | A kind of method and system for assessing user's representation data | |
CN113627361A (en) | Training method and device for face recognition model and computer program product | |
CN110033165A (en) | The recommended method of overdue loaning bill collection mode, device, medium, electronic equipment | |
CN108647714A (en) | Acquisition methods, terminal device and the medium of negative label weight | |
CN116935368A (en) | Deep learning model training method, text line detection method, device and equipment | |
CN108460049A (en) | A kind of method and system of determining information category | |
CN113051396B (en) | Classification recognition method and device for documents and electronic equipment | |
CN115359322A (en) | Target detection model training method, device, equipment and storage medium | |
CN114764874A (en) | Deep learning model training method, object recognition method and device |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20171024 |