CN101887582A - Curve corner point detection method based on difference accumulated values and three-point chain code differences - Google Patents
Curve corner point detection method based on difference accumulated values and three-point chain code differences Download PDFInfo
- Publication number
- CN101887582A CN101887582A CN 201010191596 CN201010191596A CN101887582A CN 101887582 A CN101887582 A CN 101887582A CN 201010191596 CN201010191596 CN 201010191596 CN 201010191596 A CN201010191596 A CN 201010191596A CN 101887582 A CN101887582 A CN 101887582A
- Authority
- CN
- China
- Prior art keywords
- point
- chain code
- points
- value
- curve
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The invention relates to a curve corner point detection method based on difference accumulated values and three-point chain code differences, which is technically characterized by comprising the following steps of: firstly carrying out Freeman chain code encoding to the obtained curve, calculating the difference accumulated values of the chain codes point by point according to the Freeman chain code encoding sequence based on the obtained Freeman chain codes, and dividing all pixel points on the curve into determining points, non-corner points and doubtful points through the difference accumulated values of the three-point chain codes; then calculating the three-point chain code differences of the doubtful points, and judging the doubtful points through the three-point chain code differences to find out true corner points in the doubtful points; and finally merging the corner points found in the determining points and the doubtful points to obtain all corner points of the curve. In the method, the points with curvatures to be calculated are judged by adopting the three-point chain code differences to replace the curvatures, thereby further reducing operation workload and lowering operation complexity. The corner points detected in the method also contain sequence characteristics which provide convenient for utilizing the corner points to carry out image processing.
Description
Technical field
The present invention relates to a kind of curve corner point detection method, can be used for that curve is carried out corner point and detect, belong to image characteristics extraction based on difference accumulated value and 3 chain code differences.
Background technology
In Digital Image Processing and computer vision, corner point is comprising the important information of target object.Utilize corner point information can carry out shape analysis, pattern-recognition, images match, data compression and motion analysis etc.Therefore corner point detects a research emphasis that becomes gradually in computer vision and the Flame Image Process.At present the corner point detection method roughly can be divided into two classes (1) and utilizes a mask operator to handle each pixel and neighborhood territory pixel thereof in the image, selects corner point according to result then.(2) carrying out corner point according to the marginal information of object detects.The second class algorithm generally is better than first kind algorithm, and this is that algorithm is simple relatively, is easy to realize because the second class algorithm is to carry out corner point to detect on the basis of object boundary search, and therefore second class methods become the focus of present research.
Corner point to second class methods has different descriptions, but summing up to get up detected corner point should be the point that sufficiently high point of curvature or curvature have significant change, therefore in the second class angular-point detection method, inevitably need the curvature of calculated curve.Conventional method generally adopts each curvature of calculating to detect corner point or judges corner point at present, because the point on the curve is a lot, causes conventional method detection computations amount big like this, is difficult for realizing.Also propose to get rid of earlier some non-corner points at present, the curvature value that only calculates part point then detects the method for corner point, this method can remove to exclude some non-corner points, but the number of the point that can get rid of is very limited, that have even near the pixel of primary curve, this method still needs to carry out curvature calculating simultaneously, and therefore calculated amount is still very big in actual use.The present invention proposes a kind of method of new detection corner point, this method can reduce the number that needs to calculate curvature points in a large number, compare primary curve Freeman chain code-point, the number that needs to detect curvature points can reduce by one even two orders of magnitude, for some curves commonly used, the number that need carry out the point of curvature detection can reach ten even units, has significantly reduced operand by such processing, has improved processing speed; The present invention simultaneously also adopts 3 chain code differences to replace curvature that the point that needs calculate curvature is judged, has further reduced operand, has reduced computational complexity, is more suitable for hardware and realizes.Adopt the detected corner point of this method also to comprise the ordinal characteristics of corner point at last, this provides convenience for further utilizing corner point to carry out Flame Image Process.
Summary of the invention
The technical matters that solves
For fear of the deficiencies in the prior art part, the present invention proposes a kind of curve corner point detection method based on difference accumulated value and 3 chain code differences,
Thought of the present invention is: this method is to be based upon on the basis that image curve obtained, at first carry out Freeman chain code coding to obtaining edge image, on the basis that obtains the Freeman chain code, calculate difference accumulated value, all points are divided into definite point, non-corner point and suspicious points three classes by difference accumulated value according to the pointwise of Freeman chain code coded sequence; 3 chain codes that only need to calculate these suspicious points then are poor, by 3 chain code differences suspicious points is judged, seek out the real corner point in the suspicious points, will determine a little at last and merge, promptly obtain whole corner points of curve from the true corner point that suspicious points is found out.
Technical scheme
A kind of curve corner point detection method based on difference accumulated value and 3 chain code differences is characterized in that step is as follows:
Step 1: curve is carried out all directions obtain the Freeman chain code, calculate difference accumulated value: d (i)=d according to the pointwise of Freeman chain code coded sequence to Freeman chain code coding
1(i)+d
2(i), wherein i is the index value of curve picture element, and d (i) is a difference accumulated value, d
1(i) be the difference value of consecutive point, d
2(i) be the difference value of two points in interval;
Described d
1(i), when | c (i+1)-c (i) | in the time of＜4, d
1(i)=| c (i+1)-c (i) |; When | c (i+1)-c (i) | in the time of＞4, d
1(i)=8-|c (i+1)-c (i) |; When | c (i+1)-c (i) | in the time of=4, d
1(i)=4;
Described d
2(i), when | c (i+2)-c (i-1) | in the time of＜4, d
2(i)=| c (i+2)-c (i-1) |; When | c (i+2)-c (i-1) | in the time of＞4, d
2(i)=8-|c (i+2)-c (i-1) |; When | c (i+2)-c (i-1) | in the time of=4, d
2(i)=4; Wherein c (i) is an i point chain code value;
Step 2: points all on the curve is divided into definite point, non-corner point and suspicious points according to difference accumulated value:
When i difference accumulated value d (i)＞3 that puts is corner point;
When i difference accumulated value d (i)＜3 that puts is non-corner point;
When i difference accumulated value d (i)=3 that puts is suspicious points;
Step 3: three of suspicious points chain codes are poor in the calculation procedure 2: Diff (i)=Sum (i+3)-Sum (i), wherein Diff (i) 3 chain codes of ordering for i are poor, 3 chain codes that Sum (i) is ordered for i and;
Described Sum (i)=A (i)+A (i-1)+A (i-2), the absolute chain code value of preceding two points that the absolute chain code value of the previous point that A (i) the absolute chain code value of ordering for i wherein, A (i-1) are ordered for i, A (i-2) are ordered for i, i=3,4,5,6,7
Described A (i)=A (i-1)+R (i), wherein R (i) the relative chain code value of ordering for i; A (i-1) is the absolute chain code value of the previous point of i point, i=1,2,3,4,5,6,7 ..., when i=1, A (i-1)=A (0)=0;
Described R (i)=[c (i)-c (i-1)+8] Mod8, when R (i)＜4, R (i)=R (i); When R (i)＞4, R (i)=R (i)-8; Wherein c (i) is an i point chain code value; [] Mod8 is for to carry out mould 8 computings to numerical value in the bracket or expression formula;
Step 4: 3 the chain code differences and the threshold value of suspicious points are compared, when 3 chain code differences during greater than threshold value this point be corner point, this point is non-flex point when 3 chain code differences are less than or equal to threshold value; Described threshold value is 3～6; The corner point that corner point that this step is obtained and step 3 obtain merges according to the sequencing of chain code coding, obtains all corner points of image curve.
Described curve is that width is the digitizing two-value curve of a pixel.
Beneficial effect
The curve corner point detection method based on difference accumulated value and 3 chain code differences that the present invention proposes, have three advantages: 1, calculated amount is little.Owing to only need suspicious points is carried out the judgement of curvature, and can get rid of a large amount of suspicious points and definite point by 3 chain code difference sums, the suspicious points number that obtains seldom, it is just very little to carry out the calculated amount that curvature calculates like this; 2 calculate simply.Adopt 3 chain code differences to replace curvature to calculate,, so just avoided calculating this class of curvature and need carry out the computing of floating-point and arc cosine, calculate simply, be easy to hardware and realize because 3 chain code differences only are to carry out some simple addition and subtractions; 3 practicability and effectiveness.From last experimental result, the corner point accurate and effective that this method detects.
Description of drawings
The basic flow sheet of Fig. 1 method of the present invention
Fig. 2 Freeman8 direction value
Fig. 3 Freeman chain code points to
Fig. 4 uses this method to finish the example 1 that corner point detects
(a) former figure
(b) determine the some synoptic diagram
(c) suspicious points synoptic diagram
(d) final detection result
Fig. 5 uses this method to finish the example 2 that corner point detects
(a) former figure
(b) determine the some synoptic diagram
(c) suspicious points synoptic diagram
(d) final detection result
Embodiment
Now in conjunction with the embodiments, accompanying drawing is further described the present invention:
The hardware environment that is used to implement is: Pentium-43G computing machine, 1GB internal memory, 128M video card, the software environment of operation is: Matlab7.0 and Windows XP.We have realized the method that the present invention proposes with Matlab software.The concrete enforcement of the present invention is as follows:
1.Freeman chain code Sequence Detection
Chain code is to define with eight adjacent directions that center pixel points to it, the pointing direction of chain code as shown in Figure 2:
Article one, curve forms n bar chain after by grid discretization, and this curving chain code can be expressed as { c (i) }
n, a direction in eight directions of every chain sensing, c (i) ∈ 0,1 ... 7}, i are the index value of pixel, and c (i) is the direction chain code that is pointed to pixel (i+1) by pixel (i), as shown in Figure 3.For example, if current pixel is p (i), b
7Be the next pixel on the curve, then p (i) value is 7.
Curving chain code detection order carrying out respectively according to the chain code direction according to the parity difference, when the chain code value adds 1, if its direction even number, then carry out one-time detection along direction, if do not detect curve point, after then being rotated counterclockwise 45 degree, detect according to clockwise direction along direction.If direction is an odd number, then earlier direction is rotated counterclockwise 45 degree and detects, there is not curve point if cross, continue to detect according to clockwise direction; For example, if last time, detected value was 4, then this at first detected value be whether 4 direction has curve point, the testing process that then enters down any is arranged, if do not have, then detect by 5321076 direction; If last time, detected value was 1, whether this should at first detect 2 direction curve point, and the testing process that then enters down any is arranged, as if there not being then the direction according to 1076543 detect.Such freeman Chain Code Detection method depends on last detection, this method can be avoided the scanning to all pixels on the one hand, increase the efficient that curve is followed the tracks of, the corner point that adopts the method for this differentiation odevity direction to help more carrying out below on the other hand detects.
2. curve pixel classification
For the non-corner point of exclusive segment, formula (1) (2) (3) below at first defining:
d(i)＝d
1(i)+d
2(i) (3)
I is the index value of curve picture element in the following formula, and d (i) is a difference accumulated value, d
1(i) be the difference value of neighbor, d
2(i) be the pixel difference value of two pixels in interval; After obtaining d (i), the criterion below utilizing is divided into following three classes with points all on the curve:
(1) if d (i)＞3 then thinks corner point;
(2) if d (i)＜3 then thinks it is non-corner point;
(3) if d (i)=3 thinks suspicious points this moment.
The first kind promptly thinks it must is corner point for determining point; The non-corner point of second class promptly is corner point scarcely; The 3rd class suspicious points promptly can not determine to be corner point.By only needing the 3rd class point is calculated curvature after the top judgement classification.
Corner point should be had a few and a part the 3rd class point by the first kind on the curve like this.Below the 3rd class point is done further judgement.
3. 3 chain codes are poor
In order to reduce operand, avoid calculating such floating-point operation of curvature and arc cosine computing, adopt 3 chain code differences to replace curvature that the 3rd class point is judged here.Before introducing 3 chain code differences, at first introduce the relative chain code of Freeman chain code, absolute chain code and 3 chain codes and notion.
A bit point to the chain code c1 of this point and this point before all having for borderline each point and point to any chain code c2 of back.So-called chain code relatively is meant the mutual relationship of c2 and c1, promptly when their directions are consistent, if their relative chain code is 0. c2 when being rotated counterclockwise with respect to c1, chain code is by the big or small value 1-4 of drift angle relatively, and correspondence is rotated counterclockwise 45 °, 90 °, 135 ° and 180 ° respectively; When turning clockwise, relative chain code value-1-3, correspondence turns clockwise 45 °, 90 ° and 135 ° respectively.So-called absolute chain code is meant the accumulated value that begins relative chain code from starting point, and the absolute chain code of starting point is made as 0, like this, moves along the border and a week gets back to starting point, and its absolute chain code value increases by 8.After having introduced absolute chain code notion, ask chain code and the time just need not consider the problems of value of 0 and 7 intersection chain codes again.
Absolute chain code and 3 chain codes and calculating:
If c (i) and c (i-1) are respectively the chain code of current some i and preceding 1 i-1, R (i) is their relative chain code, and A (i) and A (i-1) are respectively current point and more preceding absolute chain code, and then the computation process of absolute chain code is as follows.
A(0)＝0 (4)
R(i)＝[c(i)-c(i-1)+8]Mod8 (5)
When R (i)＜4, R (i)=R (i); When R (i)＞4, R (i)=R (i)-8 (6)
A(i)＝A(i-1)+R(i) (7)
3 chain codes and (be called for short later on chain code and), are current point and preceding absolute chain code sum at 2, promptly
Sum(i)＝A(i)+A(i-1)+A(i-2) (8)
For closed curve, initial 2 when calculating, the value of preceding millet cake will note the adjusted value 8 of absolute chain code at this moment, that is: around reaching to afterbody
Sum(0)＝A(0)+A(N-1)+A(N-2)-16 (9)
Sum(1)＝A(1)+A(0)+A(N-1)-8 (10)
Behind the Sum that so calculates (0), more all Sum (i) value is all deducted former Sum (0) value, promptly is able to 3 chain codes and the sequence of 0 beginning thus,
3 chain code differences be meant leave this point and 3 chain codes that enter this point and poor, be calculated as follows
Diff(i)-Sum(i+3)-Sum(i) (11)
3 chain code differences are represented the difference between both direction, and it is an amount that is directly proportional with curvature, and therefore here we can carry out the judgement of suspicious corner point with it.
Adopting 3 chain code differences to carry out the 3rd class point is the judgement of suspicious points, can further reduce calculated amount, and 3 chain code differences only need to carry out the simple integer plus-minus to be calculated, and has avoided complicated floating-point budget and arc cosine computing.
4. the judgement of suspicious points,
A threshold value at first is set, threshold value does not have fixing value in theory, be provided with according to the own severe degree that corner point is judged, because what calculate above is that 3 chain codes are poor, therefore 3 chain code differences are that 1 represent this curvature difference be 15 to spend, if be 3 chain code differences 3 think curvature difference be 45 the degree, if 3 chain code differences are-3 certainly, think that then curvature difference is-45 degree, here positive negative indication the concavity and convexity of corner point, when the trend of Freeman Chain Code Detection when being clockwise, 3 chain code differences are that positive point is recessed corner point, then are protruding corner point for bearing, and therefore utilize 3 chain code differences can also be easy to obtain the concavity and convexity of corner point, for the judgement of the 3rd class point, it is comparatively suitable that threshold value is generally got 3-6.Point greater than threshold value is thought corner point, otherwise thinks it is non-corner point.To determine a little at last and merge, promptly obtain whole corner points of curve from the true corner point that suspicious points is found out.
This method is based on the Freeman chain code and carries out that corner point detects, because the Freeman chain code itself is comprising ordinal characteristics, therefore the corner point that detects according to this method is also comprising ordinal characteristics, difference according to Freeman chain code coding trend, its corner point also is by being arranged in order clockwise or counterclockwise, and the information of this sequence is provided convenience for further handling.
Claims (2)
1. curve corner point detection method based on difference accumulated value and 3 chain code differences is characterized in that step is as follows:
Step 1: curve is carried out all directions obtain the Freeman chain code, calculate difference accumulated value: d (i)=d according to the pointwise of Freeman chain code coded sequence to Freeman chain code coding
1(i)+d
2(i), wherein i is the index value of curve picture element, and d (i) is a difference accumulated value, d
1(i) be the difference value of consecutive point, d
2(i) be the difference value of two points in interval;
Described d
1(i), when | c (i+1)-c (i) | in the time of＜4, d
1(i)=| c (i+1)-c (i) |; When | c (i+1)-c (i) | in the time of＞4, d
1(i)=8-|c (i+1)-c (i) |; When | c (i+1)-c (i) | in the time of=4, d
1(i)=4;
Described d
2(i), when | c (i+2)-c (i-1) | in the time of＜4, d
2(i)=| c (i+2)-c (i-1) |; When | c (i+2)-c (i-1) | in the time of＞4, d
2(i)=8-|c (i+2)-c (i-1) |; When | c (i+2)-c (i-1) | in the time of=4, d
2(i)=4; Wherein c (i) is an i point chain code letter;
Step 2: points all on the curve is divided into definite point, non-corner point and suspicious points according to difference accumulated value:
When i difference accumulated value d (i)＞3 that puts is corner point;
When i difference accumulated value d (i)＜3 that puts is non-corner point;
When i difference accumulated value d (i)=3 that puts is suspicious points;
Step 3: three of suspicious points chain codes are poor in the calculation procedure 2: Diff (i)=Sum (i+3)-Sum (i), wherein Diff (i) 3 chain codes of ordering for i are poor, 3 chain codes that Sum (i) is ordered for i and;
Described Sum (i)=A (i)+A (i-1)+A (i-2), the absolute chain code value of preceding two points that the absolute chain code value of the previous point that A (i) the absolute chain code value of ordering for i wherein, A (i-1) are ordered for i, A (i-2) are ordered for i, i=3,4,5,6,7
Described A (i)=A (i-1)+R (i), wherein R (i) the relative chain code value of ordering for i; A (i-1) is the absolute chain code value of the previous point of i point, i=1,2,3,4,5,6,7 ..., when i=1, A (i-1)=A (0)=0;
Described R (i)=[c (i)-c (i-1)+8] Mod8, when R (i)＜4, R (i)=R (i); When R (i)＞4, R (i)=R (i)-8; Wherein c (i) is an i point chain code value; [] Mod8 is for to carry out mould 8 computings to numerical value in the bracket or expression formula;
Step 4: 3 the chain code differences and the threshold value of suspicious points are compared, when 3 chain code differences during greater than threshold value this point be corner point, this point is non-flex point when 3 chain code differences are less than or equal to threshold value; Described threshold value is 3～6; The corner point that corner point that this step is obtained and step 3 obtain merges according to the sequencing of chain code coding, obtains all corner points of image curve.
2. the curve corner point detection method based on difference accumulated value and 3 chain code differences according to claim 1 is characterized in that: described curve is that width is the digitizing two-value curve of a pixel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010191596XA CN101887582B (en) | 2010-06-03 | 2010-06-03 | Curve corner point detection method based on difference accumulated values and three-point chain code differences |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010191596XA CN101887582B (en) | 2010-06-03 | 2010-06-03 | Curve corner point detection method based on difference accumulated values and three-point chain code differences |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101887582A true CN101887582A (en) | 2010-11-17 |
| CN101887582B CN101887582B (en) | 2011-12-14 |
Family
ID=43073492
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010191596XA Expired - Fee Related CN101887582B (en) | 2010-06-03 | 2010-06-03 | Curve corner point detection method based on difference accumulated values and three-point chain code differences |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101887582B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102615052A (en) * | 2012-02-21 | 2012-08-01 | 上海大学 | Machine visual identification method for sorting products with corner point characteristics |
| CN105574864A (en) * | 2015-12-14 | 2016-05-11 | 浙江工业大学 | Angle accumulation-based self-adapted corner point detection method |
| CN106780377A (en) * | 2016-12-07 | 2017-05-31 | 天津大学 | A kind of contour smoothing method based on Freeman chain codes in medical image segmentation |
| CN109711418A (en) * | 2019-01-29 | 2019-05-03 | 浙江大学 | A kind of contour corner detection method for object plane image |
| CN111462147A (en) * | 2020-04-30 | 2020-07-28 | 柳州智视科技有限公司 | Method for cutting and filling image block based on image block outer contour and angular point thereof |
| CN111462153A (en) * | 2020-04-30 | 2020-07-28 | 柳州智视科技有限公司 | Corner feature extraction method based on image contour Freeman chain code |
| CN112258540A (en) * | 2020-11-18 | 2021-01-22 | 西安邮电大学 | Image corner detection method based on nonlinear direction derivative |
| CN116728420A (en) * | 2023-08-11 | 2023-09-12 | 苏州安博医疗科技有限公司 | Mechanical arm regulation and control method and system for spinal surgery |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050207652A1 (en) * | 2004-03-19 | 2005-09-22 | Lockheed Martin Corporation | Methods and systems for automatic detection of corners of a region |
| CN101201938A (en) * | 2006-12-13 | 2008-06-18 | 上海吉量软件科技有限公司 | Filling algorithm for rapidly matching pair of left and right boundary point |
| CN101251926A (en) * | 2008-03-20 | 2008-08-27 | 北京航空航天大学 | Remote sensing image registration method based on local configuration covariance matrix |
| US20100080463A1 (en) * | 2008-09-27 | 2010-04-01 | Ningbo Sunrun Elec. & Info. Co., Ltd. | On-line identifying method of hand-written Arabic letter |
-
2010
- 2010-06-03 CN CN201010191596XA patent/CN101887582B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050207652A1 (en) * | 2004-03-19 | 2005-09-22 | Lockheed Martin Corporation | Methods and systems for automatic detection of corners of a region |
| CN101201938A (en) * | 2006-12-13 | 2008-06-18 | 上海吉量软件科技有限公司 | Filling algorithm for rapidly matching pair of left and right boundary point |
| CN101251926A (en) * | 2008-03-20 | 2008-08-27 | 北京航空航天大学 | Remote sensing image registration method based on local configuration covariance matrix |
| US20100080463A1 (en) * | 2008-09-27 | 2010-04-01 | Ningbo Sunrun Elec. & Info. Co., Ltd. | On-line identifying method of hand-written Arabic letter |
Non-Patent Citations (2)
| Title |
|---|
| 《1992 IEEE International Conference on Robotics and Automation》 19920514 Sohn, K. ctc. Curvature estimation and unique corner point detection for boundary representation 全文 1-2 第2卷, 2 * |
| 《自动化技术与应用》 20091231 汪剑 等 基于Freeman链码的汉字图像轮廓曲线拐角点检测方法 全文 1-2 第28卷, 第1期 2 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102615052B (en) * | 2012-02-21 | 2013-12-25 | 上海大学 | Machine visual identification method for sorting products with corner point characteristics |
| CN102615052A (en) * | 2012-02-21 | 2012-08-01 | 上海大学 | Machine visual identification method for sorting products with corner point characteristics |
| CN105574864B (en) * | 2015-12-14 | 2018-05-22 | 浙江工业大学 | The self-adaptive angular-point detection method to be added up based on angle |
| CN105574864A (en) * | 2015-12-14 | 2016-05-11 | 浙江工业大学 | Angle accumulation-based self-adapted corner point detection method |
| CN106780377B (en) * | 2016-12-07 | 2019-12-20 | 天津大学 | Contour smoothing method based on Freeman chain code in medical image segmentation |
| CN106780377A (en) * | 2016-12-07 | 2017-05-31 | 天津大学 | A kind of contour smoothing method based on Freeman chain codes in medical image segmentation |
| CN109711418A (en) * | 2019-01-29 | 2019-05-03 | 浙江大学 | A kind of contour corner detection method for object plane image |
| CN111462147A (en) * | 2020-04-30 | 2020-07-28 | 柳州智视科技有限公司 | Method for cutting and filling image block based on image block outer contour and angular point thereof |
| CN111462153A (en) * | 2020-04-30 | 2020-07-28 | 柳州智视科技有限公司 | Corner feature extraction method based on image contour Freeman chain code |
| CN111462147B (en) * | 2020-04-30 | 2022-07-05 | 柳州智视科技有限公司 | Method for cutting and filling image block based on image block outer contour and angular point thereof |
| CN111462153B (en) * | 2020-04-30 | 2023-05-19 | 柳州智视科技有限公司 | Corner feature extraction method based on image contour Freeman chain code |
| CN112258540A (en) * | 2020-11-18 | 2021-01-22 | 西安邮电大学 | Image corner detection method based on nonlinear direction derivative |
| CN112258540B (en) * | 2020-11-18 | 2022-04-22 | 西安邮电大学 | Image corner detection method based on nonlinear direction derivative |
| CN116728420A (en) * | 2023-08-11 | 2023-09-12 | 苏州安博医疗科技有限公司 | Mechanical arm regulation and control method and system for spinal surgery |
| CN116728420B (en) * | 2023-08-11 | 2023-11-03 | 苏州安博医疗科技有限公司 | Mechanical arm regulation and control method and system for spinal surgery |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101887582B (en) | 2011-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101887582B (en) | Curve corner point detection method based on difference accumulated values and three-point chain code differences | |
| Duan et al. | An improved Hough transform for line detection | |
| CN100595792C (en) | Vehicle detecting and tracing method based on video technique | |
| CN107704801B (en) | Curve lane line detection method based on segmented straight line and segmented Bezier curve | |
| CN109118523A (en) | A kind of tracking image target method based on YOLO | |
| CN111291603B (en) | Lane line detection method, device, system and storage medium | |
| CN102999926B (en) | A kind of image vision significance computational methods merged based on low-level image feature | |
| Fan et al. | Faster-than-real-time linear lane detection implementation using soc dsp tms320c6678 | |
| CN102938062A (en) | Document image slant angle estimation method based on content | |
| CN112906583B (en) | Lane line detection method and device | |
| WO2020107326A1 (en) | Lane line detection method, device and computer readale storage medium | |
| CN104346597A (en) | QR (quick response) code detection and correction extraction method and IP (intellectual property) core | |
| CN102509293B (en) | Method for detecting consistency of different-source images | |
| CN103034833B (en) | Bar code positioning method and bar code detection device | |
| EP2951783A1 (en) | Method and system for detecting moving objects | |
| US8249387B2 (en) | Image processing method and apparatus for detecting lines of images and start and end points of lines | |
| CN105469057A (en) | Hough line detection and projection-based automobile window detection method | |
| Meus et al. | Embedded vision system for pedestrian detection based on HOG+ SVM and use of motion information implemented in Zynq heterogeneous device | |
| CN103440785A (en) | Method for rapid lane departure warning | |
| Suetake et al. | Generalized fuzzy Hough transform for detecting arbitrary shapes in a vague and noisy image | |
| Jang et al. | Linear band detection based on the Euclidean distance transform and a new line segment extraction method | |
| CN104021564A (en) | Adaptive mean shift algorithm based on local invariant feature detection | |
| Martinez-Ortiz et al. | Curvature weighted gradient based shape orientation | |
| CN102131078B (en) | Video image correcting method and system | |
| CN104574396A (en) | Straight line detection method and system |
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: NORTHWESTERN POLYTECHNICAL UNIVERSITY Effective date: 20141125 Owner name: HAIAN XINXING CHEMICAL FIBER CO., LTD. Free format text: FORMER OWNER: NORTHWESTERN POLYTECHNICAL UNIVERSITY Effective date: 20141125 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: 710072 XI'AN, SHAANXI PROVINCE TO: 226600 NANTONG, JIANGSU PROVINCE |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20141125 Address after: 226600, Jiangsu County, Nantong City, Haian Province Ya Zhou Village 7 groups of banks Patentee after: Haian Xinxing Chemical Fiber Co., Ltd. Patentee after: Northwestern Polytechnical University Address before: 710072 Xi'an friendship West Road, Shaanxi, No. 127 Patentee before: Northwestern Polytechnical University |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20111214 Termination date: 20190603 |