CN102663384B - Curve identification method based on Bezier control point searching and apparatus thereof - Google Patents

Curve identification method based on Bezier control point searching and apparatus thereof Download PDF

Info

Publication number
CN102663384B
CN102663384B CN201210091188.6A CN201210091188A CN102663384B CN 102663384 B CN102663384 B CN 102663384B CN 201210091188 A CN201210091188 A CN 201210091188A CN 102663384 B CN102663384 B CN 102663384B
Authority
CN
China
Prior art keywords
hessian
image
reference mark
curve
point
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201210091188.6A
Other languages
Chinese (zh)
Other versions
CN102663384A (en
Inventor
李党
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Netposa Technologies Ltd
Original Assignee
Beijing Zanb Science & Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Zanb Science & Technology Co Ltd filed Critical Beijing Zanb Science & Technology Co Ltd
Priority to CN201210091188.6A priority Critical patent/CN102663384B/en
Publication of CN102663384A publication Critical patent/CN102663384A/en
Application granted granted Critical
Publication of CN102663384B publication Critical patent/CN102663384B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

Provided in the invention is a curve identification method based on Bezier control point searching. The method comprises the following steps that: gaussian smoothing processing, Hessian calculation, and Hessian image thinning processing are carried out on a current image to obtain a Hessian thinning image; according to an actual curve position, a user calibrates three control points on a video image; according to the three control points, a Bezier curve construction model is employed to obtain a Bezier curve; and according to the Bezier curve and the three control points, a detection curve on the Hessian thinning image is obtained and is output. According to the invention, detection on a curve inside a video image can be realized.

Description

Curves Recognition method and device based on the search of Bezier reference mark
Technical field
The present invention relates to image and process, particularly Curves Recognition method and device.
Background technology
Blood vessel in medical domain detects, palmmprint detects, cell detection, and interrupt in power domain etc., all need to be applied to curve tracking technique, so the research of curve tracking technique has great significance in above-mentioned field.
There is following problem in traditional curve tracking technique: 1) responsive to noise image; 2) to existing the image of background interference responsive; 3) cannot accurately identify cross curve; 4) there is the curve of fracture in None-identified; 5) for image, there is situation about changing, cannot accurately identify curve to be detected; 6) cannot identify in real time the curve existing in image; 7) once curve detection mistake does not have automatic correction mechanism identification curve.
In sum, in the urgent need to proposing, a kind of energy is simple, effectively tracking connects Curves Recognition method and the device in complex scene at present.
Summary of the invention
In view of this, fundamental purpose of the present invention is to realize and can simply, effectively identifies the curve in actual scene.
For achieving the above object, according to first aspect of the present invention, provide a kind of Curves Recognition method based on the search of Bezier reference mark, the method comprising the steps of:
First step, carries out Gaussian smoothing, Hessian calculating and the operation of Hessian image thinning to present image, obtains Hessian refined image;
Second step, user, according to actual curve position, demarcates 3 reference mark on video image;
Third step, according to 3 reference mark, adopts Bezier curve to build modelling and obtains Bezier curve;
The 4th step, obtains detection curve the output in Hessian refined image according to Bezier curve and 3 reference mark.
In wherein said first step, Gaussian smoothing is to adopt Gaussian smoothing function to present image (wherein, the image that present image receives in the time of can being true-time operation, also can be one section of current frame image in video) carry out filtering processing, to obtain current smoothed image, specific as follows: establishing present image is
Figure 314818DEST_PATH_IMAGE002
(
Figure 512056DEST_PATH_IMAGE002
be point in present image
Figure 678595DEST_PATH_IMAGE004
gray-scale value), the smoothed image obtaining after processing after filtering ( be the interior point of smoothed image of present image
Figure 991393DEST_PATH_IMAGE004
gray-scale value) be
Figure 977935DEST_PATH_IMAGE008
Wherein,
Figure 50933DEST_PATH_IMAGE010
the kernel function that represents Gaussian smoothing,
Figure 829009DEST_PATH_IMAGE012
represent convolution algorithm.The kernel function of Gaussian smoothing in
Figure 246532DEST_PATH_IMAGE014
value relevant with the width of line, meet
Figure 377299DEST_PATH_IMAGE016
, wherein
Figure 578473DEST_PATH_IMAGE018
the pixel wide that represents line.
In described first step, Hessian calculates and comprises:
Step a) is utilized every point in current smoothed image
Figure 974950DEST_PATH_IMAGE004
the second order Grad of gray-scale value build Hessian matrix , the eigenwert of calculating Hessian matrix
Figure 85306DEST_PATH_IMAGE022
, wherein,
Figure 457381DEST_PATH_IMAGE024
,
Figure 361663DEST_PATH_IMAGE026
, represent point
Figure 130216DEST_PATH_IMAGE004
with the shade of gray difference of eight adjacent side points around, its computing formula is as follows:
Figure 610875DEST_PATH_IMAGE030
Figure 635780DEST_PATH_IMAGE034
Wherein,
Figure 392384DEST_PATH_IMAGE036
represent pixel gray-scale value;
If the eigenwert of certain any Hessian matrix in step b) smoothed image
Figure 964628DEST_PATH_IMAGE040
, TH1 ∈ [4.2,4.9], thinks that this point is Hessian point, otherwise this point of filtering, thereby obtain Hessian image.
In described first step, the operation of Hessian image thinning is that Hessian image is carried out to morphologic thinning operation, to obtain Hessian refined image.
Described third step is according to 3 reference mark
Figure 169956DEST_PATH_IMAGE042
, adopting Bezier curve to build modelling and determine Bezier curve, its formula is as follows:
Figure 781066DEST_PATH_IMAGE044
。
Described the 4th step further comprises:
Step 1041, simultaneously by 3 reference mark
Figure 682157DEST_PATH_IMAGE042
Figure 11508DEST_PATH_IMAGE046
,
Figure 757878DEST_PATH_IMAGE048
,
Figure 489073DEST_PATH_IMAGE050
,
Figure 561066DEST_PATH_IMAGE052
on four direction, carry out integral translation,
Figure 112133DEST_PATH_IMAGE054
,
Figure 924844DEST_PATH_IMAGE056
, wherein k1, n are threshold parameter, k1 ∈ [4,6] and be integer, and n ∈ [2,4] and be integer, can build according to the new reference mark after integral translation the Bezier curve that 4n bar is new;
Step 1042, for every new Bezier curve, exists to the every bit on this Bezier curve respectively
Figure 510546DEST_PATH_IMAGE058
the nearest Hessian point of search in scope,
Figure 487860DEST_PATH_IMAGE060
and be integer, and be integer, add up on this Bezier curve whole points and Hessian order recently distance with;
Step 1043, the distance of more every new Bezier curve statistics and size, selected distance and minimum 3 reference mark corresponding to Bezier curve
Figure 145555DEST_PATH_IMAGE064
be 3 benchmark reference mark;
Step 1044, keeps
Figure 523446DEST_PATH_IMAGE066
two benchmark reference mark are constant, respectively with pixel in 6 pixel coverages is around interim reference mark, and according to this interim reference mark and
Figure 462900DEST_PATH_IMAGE066
build Bezier curve, calculate this Bezier length of a curve, and the number of the available point of this Bezier curve in Hessian refined image, add up on this Bezier curve the number that point belongs to Hessian refined image foreground point, and calculate the number of this available point and the ratio of this Bezier length of a curve, and if this ratio is greater than k2, k2 ∈ [0.5,0.6], using this interim reference mark as benchmark reference mark optimum controlling point ; With identical method, obtain respectively benchmark reference mark simultaneously optimum controlling point
Figure 775360DEST_PATH_IMAGE072
;
Step 1045, according to optimum controlling point
Figure 720182DEST_PATH_IMAGE074
build Bezier curve, and the also output using this Bezier curve as detection curve.
According to another aspect of the present invention, a kind of Curves Recognition square law device based on the search of Bezier reference mark is provided, this device comprises:
Hessian refined image acquiring unit, for present image being carried out to Gaussian smoothing, Hessian calculating and the operation of Hessian image thinning, obtains Hessian refined image;
Demarcation unit, 3 reference mark according to actual curve position, is demarcated 3 reference mark for user on video image;
Bezier curve acquisition unit, for according to 3 reference mark, adopts Bezier curve to build modelling and obtains Bezier curve;
Detection curve obtains and output unit, for obtaining detection curve the output in Hessian refined image according to Bezier curve and 3 reference mark.
Wherein, in described Hessian refined image acquiring unit, Gaussian smoothing is to adopt Gaussian smoothing function to present image (wherein, the image that present image receives in the time of can being true-time operation, also can be one section of current frame image in video) carry out filtering processing, to obtain current smoothed image, specific as follows: establishing present image is
Figure 885715DEST_PATH_IMAGE002
(
Figure 625001DEST_PATH_IMAGE002
be point in present image
Figure 875985DEST_PATH_IMAGE004
gray-scale value), the smoothed image obtaining after processing after filtering
Figure 562181DEST_PATH_IMAGE006
(
Figure 831489DEST_PATH_IMAGE006
be the interior point of smoothed image of present image
Figure 755058DEST_PATH_IMAGE004
gray-scale value) be
Wherein,
Figure 779963DEST_PATH_IMAGE010
the kernel function that represents Gaussian smoothing,
Figure 169356DEST_PATH_IMAGE012
represent convolution algorithm.The kernel function of Gaussian smoothing
Figure 266756DEST_PATH_IMAGE010
in
Figure 476020DEST_PATH_IMAGE014
value relevant with the width of line, meet
Figure 317068DEST_PATH_IMAGE016
, wherein
Figure 560968DEST_PATH_IMAGE018
the pixel wide that represents line.
In described Hessian refined image acquiring unit, Hessian calculates and comprises:
Step a) is utilized every point in current smoothed image
Figure 826340DEST_PATH_IMAGE004
the second order Grad of gray-scale value build Hessian matrix , the eigenwert of calculating Hessian matrix
Figure 354590DEST_PATH_IMAGE022
, wherein,
Figure 203728DEST_PATH_IMAGE024
,
Figure 892199DEST_PATH_IMAGE026
,
Figure 826788DEST_PATH_IMAGE028
represent point
Figure 6709DEST_PATH_IMAGE004
with the shade of gray difference of eight adjacent side points around, its computing formula is as follows:
Figure 975933DEST_PATH_IMAGE030
Figure 835305DEST_PATH_IMAGE032
Figure 257190DEST_PATH_IMAGE034
Wherein, represent pixel
Figure 238101DEST_PATH_IMAGE038
gray-scale value;
If the eigenwert of certain any Hessian matrix in step b) smoothed image
Figure 19107DEST_PATH_IMAGE040
, TH1 ∈ [4.2,4.9], thinks that this point is Hessian point, otherwise this point of filtering, thereby obtain Hessian image.
In described Hessian refined image acquiring unit, the operation of Hessian image thinning is that Hessian image is carried out to morphologic thinning operation, to obtain Hessian refined image.
In described Bezier curve acquisition unit according to 3 reference mark
Figure 177555DEST_PATH_IMAGE042
, adopting Bezier curve to build modelling and determine Bezier curve, its formula is as follows:
Figure 761596DEST_PATH_IMAGE044
。
Described detection curve obtains with output unit specifically for realizing following steps:
Step 1041, simultaneously by 3 reference mark
Figure 844270DEST_PATH_IMAGE046
,
Figure 224435DEST_PATH_IMAGE048
,
Figure 552780DEST_PATH_IMAGE050
,
Figure 600370DEST_PATH_IMAGE052
on four direction, carry out integral translation,
Figure 723178DEST_PATH_IMAGE054
, , wherein k1, n are threshold parameter, k1 ∈ [4,6] and be integer, and n ∈ [2,4] and be integer, can build according to the new reference mark after integral translation the Bezier curve that 4n bar is new;
Step 1042, for every new Bezier curve, exists to the every bit on this Bezier curve respectively
Figure 477604DEST_PATH_IMAGE058
the nearest Hessian point of search in scope,
Figure 317384DEST_PATH_IMAGE060
and be integer,
Figure 860361DEST_PATH_IMAGE062
and be integer, add up on this Bezier curve whole points and Hessian order recently distance with;
Step 1043, the distance of more every new Bezier curve statistics and size, selected distance and minimum 3 reference mark corresponding to Bezier curve
Figure 231430DEST_PATH_IMAGE064
be 3 benchmark reference mark;
Step 1044, keeps
Figure 885265DEST_PATH_IMAGE066
two benchmark reference mark are constant, respectively with
Figure 658180DEST_PATH_IMAGE068
pixel in 6 pixel coverages is around interim reference mark, and according to this interim reference mark and
Figure 372059DEST_PATH_IMAGE066
build Bezier curve, calculate this Bezier length of a curve, and the number of the available point of this Bezier curve in Hessian refined image, add up on this Bezier curve the number that point belongs to Hessian refined image foreground point, and calculate the number of this available point and the ratio of this Bezier length of a curve, and if this ratio is greater than k2, k2 ∈ [0.5,0.6], using this interim reference mark as benchmark reference mark
Figure 964845DEST_PATH_IMAGE068
optimum controlling point ; With identical method, obtain respectively benchmark reference mark simultaneously
Figure 46863DEST_PATH_IMAGE066
optimum controlling point
Figure 931642DEST_PATH_IMAGE072
;
Step 1045, according to optimum controlling point
Figure 11725DEST_PATH_IMAGE074
build Bezier curve, and the also output using this Bezier curve as detection curve.
According to Curves Recognition method and the device based on the search of Bezier reference mark of the present invention, can identify simply, exactly the curve in video image.
Accompanying drawing explanation
Fig. 1 shows according to the process flow diagram of the Curves Recognition method based on the search of Bezier reference mark of the present invention;
Fig. 2 shows according to the frame diagram of the Curves Recognition device based on the search of Bezier reference mark of the present invention.
Embodiment
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with embodiment and accompanying drawing, the present invention is described in more detail.
Fig. 1 represents according to the process flow diagram of the Curves Recognition method based on the search of Bezier reference mark of the present invention.As shown in Figure 1, according to the Curves Recognition method based on the search of Bezier reference mark of the present invention, comprise:
First step 101, carries out Gaussian smoothing, Hessian calculating and the operation of Hessian image thinning to present image, obtains Hessian refined image;
Second step 102, user, according to actual curve position, demarcates 3 reference mark on video image;
Third step 103, according to 3 reference mark, adopts Bezier curve to build modelling and obtains Bezier curve;
The 4th step 104, obtains detection curve the output in Hessian refined image according to Bezier curve and 3 reference mark.
first step:
Described Gaussian smoothing is to adopt Gaussian smoothing function to present image (wherein, the image that present image receives in the time of can being true-time operation, also can be one section of current frame image in video) carry out filtering processing, to obtain current smoothed image, specific as follows: establishing present image is
Figure 210625DEST_PATH_IMAGE002
, be point in present image
Figure 748234DEST_PATH_IMAGE004
gray-scale value, the smoothed image obtaining after processing after filtering
Figure 564880DEST_PATH_IMAGE006
( be the interior point of smoothed image of present image gray-scale value) be
Wherein,
Figure 244550DEST_PATH_IMAGE010
the kernel function that represents Gaussian smoothing,
Figure 785253DEST_PATH_IMAGE012
represent convolution algorithm.The kernel function of Gaussian smoothing
Figure 976194DEST_PATH_IMAGE010
in
Figure 639256DEST_PATH_IMAGE014
value relevant with the width of line, meet
Figure 915648DEST_PATH_IMAGE016
, wherein
Figure 135408DEST_PATH_IMAGE018
the pixel wide that represents line.
Hessian calculates and comprises:
Step a) is utilized every point in current smoothed image
Figure 430123DEST_PATH_IMAGE004
the second order Grad of gray-scale value build Hessian matrix
Figure 11889DEST_PATH_IMAGE020
, the eigenwert of calculating Hessian matrix
Figure 24845DEST_PATH_IMAGE022
, wherein,
Figure 907350DEST_PATH_IMAGE024
,
Figure 72883DEST_PATH_IMAGE026
,
Figure 77748DEST_PATH_IMAGE028
represent point
Figure 328732DEST_PATH_IMAGE004
with the shade of gray difference of eight adjacent side points around, its computing formula is as follows:
Figure 77245DEST_PATH_IMAGE030
Figure 97285DEST_PATH_IMAGE032
Figure 273052DEST_PATH_IMAGE034
Wherein,
Figure 8402DEST_PATH_IMAGE036
represent pixel gray-scale value;
If the eigenwert of certain any Hessian matrix in step b) smoothed image
Figure 622103DEST_PATH_IMAGE040
, TH1 ∈ [4.2,4.9], thinks that this point is Hessian point, otherwise this point of filtering, thereby obtain Hessian image.
The operation of Hessian image thinning is that Hessian image is carried out to morphologic thinning operation, to obtain Hessian refined image.Wherein, morphologic thinning operation can be with reference to " Digital Image Processing, Paul Gonzales, Electronic Industry Press ".
third step:
Described third step is according to 3 reference mark
Figure 453924DEST_PATH_IMAGE042
, adopting Bezier curve to build modelling and determine Bezier curve, its formula is as follows:
Figure 928768DEST_PATH_IMAGE044
。
the 4th step:
Described the 4th step further comprises:
Step 1041, simultaneously by 3 reference mark
Figure 832133DEST_PATH_IMAGE042
,
Figure 344334DEST_PATH_IMAGE048
,
Figure 788697DEST_PATH_IMAGE050
, on four direction, carry out integral translation, ,
Figure 157995DEST_PATH_IMAGE056
, wherein k1, n are threshold parameter, k1 ∈ [4,6] and be integer, and n ∈ [2,4] and be integer, can build according to the new reference mark after integral translation the Bezier curve that 4n bar is new;
Step 1042, for every new Bezier curve, exists to the every bit on this Bezier curve respectively
Figure 341852DEST_PATH_IMAGE058
the nearest Hessian point of search in scope,
Figure 524703DEST_PATH_IMAGE060
and be integer,
Figure 743195DEST_PATH_IMAGE062
and be integer, add up on this Bezier curve whole points and Hessian order recently distance with;
Step 1043, the distance of more every new Bezier curve statistics and size, selected distance and minimum 3 reference mark corresponding to Bezier curve
Figure 353299DEST_PATH_IMAGE064
be 3 benchmark reference mark;
Step 1044, keeps
Figure 24451DEST_PATH_IMAGE066
two benchmark reference mark are constant, respectively with
Figure 742484DEST_PATH_IMAGE068
pixel in 6 pixel coverages is around interim reference mark, and according to this interim reference mark and
Figure 753165DEST_PATH_IMAGE066
build Bezier curve, calculate this Bezier length of a curve, and the number of the available point of this Bezier curve in Hessian refined image, add up on this Bezier curve the number that point belongs to Hessian refined image foreground point, and calculate the number of this available point and the ratio of this Bezier length of a curve, and if this ratio is greater than k2, k2 ∈ [0.5,0.6], using this interim reference mark as benchmark reference mark optimum controlling point
Figure 692620DEST_PATH_IMAGE070
; With identical method, obtain respectively benchmark reference mark simultaneously
Figure 466541DEST_PATH_IMAGE066
optimum controlling point
Figure 144778DEST_PATH_IMAGE072
;
Step 1045, according to optimum controlling point
Figure 345952DEST_PATH_IMAGE074
build Bezier curve, and the also output using this Bezier curve as detection curve.
Preferably, when power transmission line curve detection, can choose k1=4, n=3, k2=0.55.
Fig. 2 shows according to the frame diagram of the Curves Recognition device based on the search of Bezier reference mark of the present invention.As shown in Figure 2, according to the Curves Recognition device based on the search of Bezier reference mark of the present invention, comprise:
Hessian refined image acquiring unit 1, for present image being carried out to Gaussian smoothing, Hessian calculating and the operation of Hessian image thinning, obtains Hessian refined image;
Demarcation unit, 3 reference mark 2 according to actual curve position, is demarcated 3 reference mark for user on video image;
Bezier curve acquisition unit 3, for according to 3 reference mark, adopts Bezier curve to build modelling and obtains Bezier curve;
Detection curve obtains and output unit 4, for obtaining detection curve the output in Hessian refined image according to Bezier curve and 3 reference mark.
Wherein, in described Hessian refined image acquiring unit 1, Gaussian smoothing is to adopt Gaussian smoothing function to present image (wherein, the image that present image receives in the time of can being true-time operation, also can be one section of current frame image in video) carry out filtering processing, to obtain current smoothed image, specific as follows: establishing present image is
Figure 742429DEST_PATH_IMAGE002
,
Figure 320041DEST_PATH_IMAGE002
be point in present image
Figure 849855DEST_PATH_IMAGE004
gray-scale value, the smoothed image obtaining after processing after filtering (
Figure 292655DEST_PATH_IMAGE006
be the interior point of smoothed image of present image
Figure 159111DEST_PATH_IMAGE004
gray-scale value) be
Figure 61208DEST_PATH_IMAGE008
Wherein,
Figure 354917DEST_PATH_IMAGE010
the kernel function that represents Gaussian smoothing, represent convolution algorithm.The kernel function of Gaussian smoothing
Figure 379822DEST_PATH_IMAGE010
in
Figure 136425DEST_PATH_IMAGE014
value relevant with the width of line, meet
Figure 887123DEST_PATH_IMAGE016
, wherein
Figure 729177DEST_PATH_IMAGE018
the pixel wide that represents line.
In described Hessian refined image acquiring unit 1, Hessian calculates and comprises:
Step a) is utilized every point in current smoothed image
Figure 937435DEST_PATH_IMAGE004
the second order Grad of gray-scale value build Hessian matrix
Figure 814125DEST_PATH_IMAGE020
, the eigenwert of calculating Hessian matrix
Figure 449636DEST_PATH_IMAGE022
, wherein,
Figure 778987DEST_PATH_IMAGE024
, ,
Figure 522132DEST_PATH_IMAGE028
represent point
Figure 325615DEST_PATH_IMAGE004
with the shade of gray difference of eight adjacent side points around, its computing formula is as follows:
Figure 79944DEST_PATH_IMAGE030
Figure 879273DEST_PATH_IMAGE032
Figure 215708DEST_PATH_IMAGE034
Wherein,
Figure 707869DEST_PATH_IMAGE036
represent pixel
Figure 762544DEST_PATH_IMAGE038
gray-scale value;
If the eigenwert of certain any Hessian matrix in step b) smoothed image
Figure 99984DEST_PATH_IMAGE040
, TH1 ∈ [4.2,4.9], thinks that this point is Hessian point, otherwise this point of filtering, thereby obtain Hessian image.
In described Hessian refined image acquiring unit 1, the operation of Hessian image thinning is that Hessian image is carried out to morphologic thinning operation, to obtain Hessian refined image.Wherein, morphologic thinning operation can be with reference to " Digital Image Processing, Paul Gonzales, Electronic Industry Press ".
In described Bezier curve acquisition unit 3 according to 3 reference mark
Figure 556504DEST_PATH_IMAGE042
, adopting Bezier curve to build modelling and determine Bezier curve, its formula is as follows:
Figure 953988DEST_PATH_IMAGE044
。
Described detection curve obtains with output unit 4 specifically for realizing following steps:
Step 1041, simultaneously by 3 reference mark
Figure 634160DEST_PATH_IMAGE046
,
Figure 945187DEST_PATH_IMAGE048
,
Figure 716834DEST_PATH_IMAGE050
,
Figure 542839DEST_PATH_IMAGE052
on four direction, carry out integral translation, ,
Figure 653194DEST_PATH_IMAGE056
, wherein k1, n are threshold parameter, k1 ∈ [4,6] and be integer, and n ∈ [2,4] and be integer, can build according to the new reference mark after integral translation the Bezier curve that 4n bar is new;
Step 1042, for every new Bezier curve, exists to the every bit on this Bezier curve respectively
Figure 658059DEST_PATH_IMAGE058
the nearest Hessian point of search in scope,
Figure 906113DEST_PATH_IMAGE060
and be integer,
Figure 389047DEST_PATH_IMAGE062
and be integer, add up on this Bezier curve whole points and Hessian order recently distance with;
Step 1043, the distance of more every new Bezier curve statistics and size, selected distance and minimum 3 reference mark corresponding to Bezier curve
Figure 674666DEST_PATH_IMAGE064
be 3 benchmark reference mark;
Step 1044, keeps
Figure 663482DEST_PATH_IMAGE066
two benchmark reference mark are constant, respectively with
Figure 588713DEST_PATH_IMAGE068
pixel in 6 pixel coverages is around interim reference mark, and according to this interim reference mark and build Bezier curve, calculate this Bezier length of a curve, and the number of the available point of this Bezier curve in Hessian refined image, add up on this Bezier curve the number that point belongs to Hessian refined image foreground point, and calculate the number of this available point and the ratio of this Bezier length of a curve, and if this ratio is greater than k2, k2 ∈ [0.5,0.6], using this interim reference mark as benchmark reference mark
Figure 812201DEST_PATH_IMAGE068
optimum controlling point
Figure 906671DEST_PATH_IMAGE070
; With identical method, obtain respectively benchmark reference mark simultaneously
Figure 381515DEST_PATH_IMAGE066
optimum controlling point
Figure 222563DEST_PATH_IMAGE072
;
Step 1045, according to optimum controlling point
Figure 466463DEST_PATH_IMAGE074
build Bezier curve, and the also output using this Bezier curve as detection curve.
Preferably, when power transmission line curve detection, can choose k1=4, n=3, k2=0.55.
According to Curves Recognition method and the device based on the search of Bezier reference mark of the present invention, can identify simply, exactly the curve in video image.
The above; be only preferred embodiment of the present invention, be not intended to limit protection scope of the present invention, be to be understood that; the present invention is not limited to implementation as described herein, and the object that these implementations are described is to help those of skill in the art to put into practice the present invention.Any those of skill in the art are easy to be further improved without departing from the spirit and scope of the present invention and perfect, therefore the present invention is only subject to the restriction of content and the scope of the claims in the present invention, and its intention contains all alternatives and equivalents that are included in the spirit and scope of the invention being limited by claims.

Claims (8)

1. the Curves Recognition method based on the search of Bezier reference mark, is characterized in that, the method comprises:
First step, carries out Gaussian smoothing, Hessian calculating and the operation of Hessian image thinning to present image, obtains Hessian refined image;
Second step, user, according to actual curve position, demarcates 3 reference mark on video image;
Third step, according to 3 reference mark, adopts Bezier curve to build modelling and obtains Bezi er curve;
The 4th step, obtains detection curve the output in Hessian refined image according to Bezier curve and 3 reference mark;
Wherein, described third step is according to 3 reference mark P 0(x 0, y 0), P 1(x 1, y 1), P 2(x 2, y 2), adopting Bezier curve to build modelling and determine Bezier curve, its formula is as follows:
P(t)＝(1-t) 2P 0+2(1-t)tP 1+t 2P 2 0≤t≤1；
Described the 4th step further comprises:
Step 1041, simultaneously by 3 reference mark P 0(x 0, y 0), P 1(x 1, y 1), P 2(x 2, y 2) at (x-Δ x, y-Δ y), (x-Δ x, y+ Δ y), (x+ Δ x, y-Δ y), (x+ Δ x, y+ Δ y) on four direction, carry out integral translation, Δ x ∈ { k1,2k1,, k1 * n}, Δ y ∈ { k1,2k1 ..., k1 * n}, wherein k1, n are threshold parameter, k1 ∈ [4,6] and be integer, n ∈ [2,4] and be integer, builds the new Bezier curve of 4n bar according to the new reference mark after integral translation;
Step 1042, respectively for every new Bezier curve, to the every bit on this Bezier curve at (Δ x ', Δ y ') the nearest Hessian point of search in scope, Δ x ' ∈ [k1, k1] and be integer, Δ y ' ∈ [k1, k1] and be integer, add up on this Bezier curve whole points and Hessian order recently distance with;
Step 1043, the distance of more every new Bezier curve statistics and size, selected distance and minimum 3 reference mark P corresponding to Bezier curve 0' (x 0', y 0'), P 1' (x 1', y 1'), P 2' (x 2', y 2') be 3 benchmark reference mark;
Step 1044, keeps P 1', P 2' two benchmark reference mark are constant, respectively with P 0pixel in ' 6 pixel coverages is around interim reference mark, and according to this interim reference mark and P 1', P 2' structure Bezier curve, calculate this Bezier length of a curve, and the number of the available point of this Bezier curve in Hes sian refined image, add up on this Bezier curve the number that point belongs to Hessi an refined image foreground point, and calculate the number of this available point and the ratio of this Bezier length of a curve, and if this ratio is greater than k2, k2 ∈ [0.5,0.6], using this interim reference mark as benchmark reference mark P 0' optimum controlling point P 0 *; With identical method, obtain respectively benchmark reference mark P simultaneously 1', P 2' optimum controlling point P 1 *, P 2 *;
Step 1045, according to optimum controlling point P 0 +, P 1 +, P 2 +build Bezier curve, and the also output using this B ezier curve as detection curve.
2. the method for claim 1, in described first step, Gaussian smoothing is specific as follows: establishing present image is I (x, y), the smoothed image I obtaining after processing after filtering σ(x, y) is
I σ(x,y)＝I(x,y)*g σ(x,y)
Wherein, I (x, y) is the gray-scale value of point (x, y) in present image, I σ(x, y) is the gray-scale value of the interior point (x, y) of smoothed image of present image, g σ(x, y) represents the kernel function of Gaussian smoothing, and * represents convolution algorithm, the kernel function g of Gaussian smoothing σvalue of σ is relevant with the width of line in (x, y), meet σ>=ω/ , wherein ω represents the pixel wide of line.
3. the method for claim 1, in described first step, Hessian calculates and comprises:
Step a) utilizes the second order Grad of the gray-scale value of every point (x, y) in current smoothed image to build Hessian matrix H ( x , y ) = r xx ( x , y ) r xy ( x , y ) r xy ( x , y ) r yy ( x , y ) , Calculate the eigenvalue λ of Hessian matrix, wherein, r xx(x, y), r xy(x, y), r yy(x, y) expression point (x, y) and around the shade of gray difference of eight adjacent side points, its computing formula is as follows:
r xx(x,y)＝|I(x-1,y)+I(x+1,y)-I(x,y)|
r yy(x,y)＝|I(x,y-1)+I(x,y+1)-I(x,y)|
r xy(x,y)＝|I(x-1,y-1)+I(x+1,y+1)-I(x-1,y+1)-I(x+1,y-1)|
Wherein, I () represents the gray-scale value of pixel ();
If the eigenvalue λ >=TH1 of certain any Hessian matrix in step b) smoothed image, T H1 ∈ [4.2,4.9], thinks that this point is Hessian point, otherwise this point of filtering, thereby obtain Hessian image.
4. the method for claim 1, in described first step, the operation of Hessian image thinning is that Hessian image is carried out to morphologic thinning operation, to obtain Hessian refined image.
5. the Curves Recognition device based on the search of Bezier reference mark, is characterized in that, this device comprises:
Hessian refined image acquiring unit, for present image being carried out to Gaussian smoothing, Hessian calculating and the operation of Hessian image thinning, obtains Hessian refined image;
Demarcation unit, 3 reference mark according to actual curve position, is demarcated 3 reference mark for user on video image;
Bezier curve acquisition unit, for according to 3 reference mark, adopts Bezier curve to build modelling and obtains Bezier curve;
Detection curve obtains and output unit, for obtaining detection curve the output in Hessian refined image according to Bezier curve and 3 reference mark;
Wherein, in described Bezier curve acquisition unit according to 3 reference mark P 0(x 0, y 0), P 1(x 1, y 1), P 2(x 2, y 2), adopting Bezier curve to build modelling and determine Bezier curve, its formula is as follows:
P(t)＝(1-t) 2P 0+2(1-t)tP 1+t 2P 2 0≤t≤1；
Described detection curve obtains with output unit specifically for realizing following steps:
Step 1041, simultaneously by 3 reference mark P 0(x 0, y 0), P 1(x 1, y 1), P 2(x 2, y 2) at (x-Δ x, y-Δ y), (x-Δ x, y+ Δ y), (x+ Δ x, y-Δ y), (x+ Δ x, y+ Δ y) on four direction, carry out integral translation, Δ x ∈ { k1,2k1,, k1 * n}, Δ y ∈ { k1,2k1 ..., k1 * n}, wherein k1, n are threshold parameter, k1 ∈ [4,6] and be integer, n ∈ [2,4] and be integer, builds the new Bezier curve of 4n bar according to the new reference mark after integral translation;
Step 1042, respectively for every new Bezier curve, to the every bit on this Bezier curve at (Δ x ', Δ y ') the nearest Hessian point of search in scope, Δ x ' ∈ [k1, k1] and be integer, Δ y ' ∈ [k1, k1] and be integer, add up on this Bezier curve whole points and Hessia n order recently distance with;
Step 1043, the distance of more every new Bezier curve statistics and size, selected distance and minimum 3 reference mark P corresponding to Bezier curve 0' (x 0', y 0'), P 1' (x 1', y 1'), P 2' (x 2', y 2') be 3 benchmark reference mark;
Step 1044, keeps P 1', P 2' two benchmark reference mark are constant, respectively with P 0pixel in ' 6 pixel coverages is around interim reference mark, and according to this interim reference mark and P 1', P 2' structure Bezier curve, calculate this Bezier length of a curve, and the number of the available point of this Bezier curve in Hes sian refined image, add up on this Bezier curve the number that point belongs to Hessi an refined image foreground point, and calculate the number of this available point and the ratio of this Bezier length of a curve, and if this ratio is greater than k2, k2 ∈ [0.5,0.6], using this interim reference mark as benchmark reference mark P 0' optimum controlling point P 0 *; With identical method, obtain respectively benchmark reference mark P simultaneously 1', P 2' optimum controlling point P 1 *, P 2 *;
Step 1045, according to optimum controlling point P 0 +, P 1 +, P 2 +build Bezier curve, and the also output using this B ezier curve as detection curve.
6. device as claimed in claim 5, is characterized in that, in described Hessian refined image acquiring unit, Gaussian smoothing is specific as follows: establishing present image is I (x, y), the smoothed image I obtaining after processing after filtering σ(x, y) is
I σ(x,y)＝I(x,y)*g σ(x,y)
Wherein, I (x, y) is the gray-scale value of point (x, y) in present image, I σ(x, y) is the gray-scale value of the interior point (x, y) of smoothed image of present image, g σ(x, y) represents the kernel function of Gaussian smoothing, and * represents convolution algorithm; The kernel function g of Gaussian smoothing σvalue of σ is relevant with the width of line in (x, y), meet σ>=ω/
Figure FDA0000365323290000042
, wherein ω represents the pixel wide of line.
7. device as claimed in claim 5, is characterized in that, in described Hessian refined image acquiring unit, Hessian calculates and comprises:
Step a) utilizes the second order Grad of the gray-scale value of every point (x, y) in current smoothed image to build Hessian matrix H ( x , y ) = r xx ( x , y ) r xy ( x , y ) r xy ( x , y ) r yy ( x , y ) , Calculate the eigenvalue λ of Hessian matrix, wherein, r xx(x, y), r xy(x, y), r yy(x, y) expression point (x, y) and around the shade of gray difference of eight adjacent side points, its computing formula is as follows:
r xx(x,y)＝|I(x-1,y)+I(x+1,y)-I(x,y)|
r yy(x,y)＝|I(x,y-1)+I(x,y+1)-I(x,y)|
r xy(x,y)＝|I(x-1,y-1)+I(x+1,y+1)-I(x-1,y+1)-I(x+1,y-1)|
Wherein, I () represents the gray-scale value of pixel ();
If the eigenvalue λ >=TH1 of certain any Hessian matrix in step b) smoothed image, T H1 ∈ [4.2,4.9], thinks that this point is Hessian point, otherwise this point of filtering, thereby obtain H essian image.
8. device as claimed in claim 5, is characterized in that, in described Hessian refined image acquiring unit, the operation of Hessian image thinning is that Hessian image is carried out to morphologic thinning operation, to obtain Hessian refined image.
CN201210091188.6A 2012-03-31 2012-03-31 Curve identification method based on Bezier control point searching and apparatus thereof Active CN102663384B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201210091188.6A CN102663384B (en) 2012-03-31 2012-03-31 Curve identification method based on Bezier control point searching and apparatus thereof

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201210091188.6A CN102663384B (en) 2012-03-31 2012-03-31 Curve identification method based on Bezier control point searching and apparatus thereof

Publications (2)

Publication Number Publication Date
CN102663384A CN102663384A (en) 2012-09-12
CN102663384B true CN102663384B (en) 2014-01-22

Family

ID=46772868

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201210091188.6A Active CN102663384B (en) 2012-03-31 2012-03-31 Curve identification method based on Bezier control point searching and apparatus thereof

Country Status (1)

Country Link
CN (1) CN102663384B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9401034B2 (en) * 2013-04-30 2016-07-26 Microsoft Technology Licensing, Llc Tessellation of two-dimensional curves using a graphics pipeline
CN103971101A (en) * 2014-05-21 2014-08-06 国家电网公司 Curve identification method based on Bethel control point searching
CN104462715B (en) * 2014-12-24 2017-09-15 黑龙江科技大学 A kind of modeling method of the photovoltaic cell output characteristic based on Bezier functions
CN105046705B (en) * 2015-07-13 2017-07-25 浙江工业大学 A kind of crack edge detection method based on fuzzy theory
US9741133B2 (en) * 2015-09-29 2017-08-22 Adobe Systems Incorporated Identifying shapes in an image by comparing Bézier curves
CN106600663B (en) * 2016-12-30 2019-07-09 浙江工业大学 A kind of simple curve smoothing method based on quadratic b-spline curve
CN107907048A (en) * 2017-06-30 2018-04-13 长沙湘计海盾科技有限公司 A kind of binocular stereo vision method for three-dimensional measurement based on line-structured light scanning
CN113313724B (en) * 2021-05-27 2022-04-08 深圳企业云科技股份有限公司 Line detection processing method for resisting resampling of mobile phone camera

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1449638A (en) * 2000-08-29 2003-10-15 迪吉塔亚丽雅有限公司 Method and system for processing moving picture data based upon vector graphic on wireless network
CN101393644B (en) * 2008-08-15 2010-08-04 华中科技大学 Hepatic portal vein tree modeling method and system thereof
US8248419B2 (en) * 2009-04-21 2012-08-21 Sony Corporation Entertainment Inc. Generation of cubic Bezier control points in computer graphics systems

Also Published As

Publication number Publication date
CN102663384A (en) 2012-09-12

Similar Documents

Publication Publication Date Title
CN102663384B (en) Curve identification method based on Bezier control point searching and apparatus thereof
CN109559310B (en) Power transmission and transformation inspection image quality evaluation method and system based on significance detection
CN105023014B (en) A kind of shaft tower target extraction method in unmanned plane inspection transmission line of electricity image
DE112016003118B4 (en) Point-to-point distance measurements in 3D camera images
CN105405142A (en) Edge defect detection method and system for glass panel
CN103927717A (en) Depth image recovery method based on improved bilateral filters
CN113378712B (en) Training method of object detection model, image detection method and device thereof
CN113379813A (en) Training method and device of depth estimation model, electronic equipment and storage medium
CN108090486A (en) Image processing method and device in a kind of game of billiards
CN108009547A (en) A kind of nameplate recognition methods of substation equipment and device
CN103674975A (en) Visual LCD (Liquid Crystal Display) Mura defect machine inspection method based on Gabor filtering and C-V (Chan-Vese) model segmentation
CN110969158A (en) Target detection method, system and device based on underwater operation robot vision
WO2023078285A1 (en) Moire pattern removal method and apparatus for text image, and electronic device
CN102073872A (en) Image-based method for identifying shape of parasite egg
CN104376323B (en) A kind of method and device for determining target range
CN103971101A (en) Curve identification method based on Bethel control point searching
CN103313068A (en) White balance corrected image processing method and device based on gray edge constraint gray world
CN110197483A (en) Deep basal pit crack detection method based on vision signal
CN108985307B (en) Water body extraction method and system based on remote sensing image
CN109598692A (en) A kind of defect inspection method based on the detection of local contrast salient region
CN107133965A (en) One kind is based on computer graphic image morphological image segmentation method
CN110930393A (en) Chip material pipe counting method, device and system based on machine vision
CN106231249A (en) A kind of unmanned derrick car
CN103714527A (en) Cutting method of digital radiography (DR) image
CN105787963A (en) Video target tracking method and device

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
ASS Succession or assignment of patent right

Owner name: NETPOSA TECHNOLOGIES, LTD.

Free format text: FORMER OWNER: BEIJING ZANB SCIENCE + TECHNOLOGY CO., LTD.

Effective date: 20150716

C41 Transfer of patent application or patent right or utility model
TR01 Transfer of patent right

Effective date of registration: 20150716

Address after: 100102, Beijing, Chaoyang District, Tong Tung Street, No. 1, Wangjing SOHO tower, two, C, 26 floor

Patentee after: NETPOSA TECHNOLOGIES, Ltd.

Address before: 100048 Beijing city Haidian District Road No. 9, building 4, 5 layers of international subject

Patentee before: Beijing ZANB Technology Co.,Ltd.

PP01 Preservation of patent right

Effective date of registration: 20220726

Granted publication date: 20140122

PP01 Preservation of patent right