CN104282027A - Circle detecting method based on Hough transformation - Google Patents
Circle detecting method based on Hough transformation Download PDFInfo
- Publication number
- CN104282027A CN104282027A CN201410594409.0A CN201410594409A CN104282027A CN 104282027 A CN104282027 A CN 104282027A CN 201410594409 A CN201410594409 A CN 201410594409A CN 104282027 A CN104282027 A CN 104282027A
- Authority
- CN
- China
- Prior art keywords
- circle
- radius
- totalizer
- center
- chained list
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/149—Segmentation; Edge detection involving deformable models, e.g. active contour models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a circle detecting method based on the Hough transformation. The circle detecting method based on the Hough transformation is characterized by comprising the steps that firstly, an eight-connected edge point chain in an image is extracted; secondly, point extraction is conducted on the edge chain which meets the point number condition multiple times, and circle parameters corresponding to the edge chain are calculated; the circle parameter with high occurrence frequency is screened out from an obtained circle parameter set, and then circle center voting is conducted according to radiuses with the gradient information within a certain range; a candidate circle center meeting conditions is screened out through an obtained circle center accumulator, the distances between all edge points and the candidate circle center are calculated, and then the radius which has the highest occurrence frequency and meets the conditions is taken as the radius of the circle. According to the circle detecting method based on the Hough transformation, the radius value of the circle is estimated by means of the sampled points, voting is accelerated by means of the gradient information within the narrow neighborhood range of the obtained radius value, in this way, the amount of calculation is effectively reduced, and the probability of obtaining a proper circle through detection is increased.
Description
Technical field
The present invention relates to Digital Image Processing, area of pattern recognition, particularly a kind of circle detection method based on Hough transform.
Background technology
In the various fields such as pattern-recognition, computer vision and reverse-engineering, usually need the circular pattern in scene or image to detect, identify the object such as localizing objects and vector quantization to reach.Detect that circle and parameter thereof are important research contents in computer vision and pattern-recognition quickly and accurately.Circle detection method common at present has the loop truss algorithm based on loop integral, the circle detection method based on genetic algorithm, based on there is the circle detection method of probability and the circle detection method etc. based on Hough transform.
Wherein, standard Hough transform (SHT) detects circle is the most traditional a kind of detection algorithm.Its great advantage is: to insensitive for noise, can effectively denoising after detection; And in circle distortion, when even subregion is lost, still more satisfactory result can be obtained.The basic thought of Hough transform is that the pixel in image space with certain relation is carried out cluster. find the parameter space accumulation corresponding point that a certain analytical form of these pixels can be connected.But because circle has 3 free parameters, need accumulate in three-dimensional parameter space, make this way because of calculated amount and memory demand excessive and do not conform to reality.
In order to reduce internal memory and elapsed time, the dimension that counting and reducing totalizer reducing to participate in Hough transform is as much as possible crucial.If any scholar, random Hough transformation method is proposed, utilize stochastic sampling to 3 of not conllinear calculate round parameter, memory consumption can be reduced largely, this algorithm performance when processing simple image is fine, but when processing complicated image, stochastic sampling introduces a large amount of invalid samplings and accumulation, makes the hydraulic performance decline of algorithm.There is the shade of gray information utilizing boundary pixel again, estimate the angle of marginal point, utilize angle information to retrain the direction of ballot, thus reduce operand.This algorithm is compared with standard Hough transform, and antijamming capability declines to some extent, and target circle the to be asked target cumulative sum in totalizer reduces greatly, easily causes the maximum value formed with noise spot close, is even surmounted, thus the judgement made the mistake.
Summary of the invention
The object of the present invention is to provide a kind of method utilizing Hough transform voting mechanism to carry out the circle in detected image, to solve the detection identification Problems existing of circle in normal image.
For achieving the above object, technical scheme of the present invention is: a kind of circle detection method based on Hough transform, is characterized in that, realizes as follows:
S1: after filtering and noise reduction is carried out to the gray level image of input, with the edge of this gray level image of Canny operator extraction, and the gradient vector of edge calculation point, then edge carries out the Contour extraction of 8 connections, obtain n bar 8 and be communicated with boundary chain C [n], n be more than or equal to 1 positive integer; Meanwhile, build one with the identical two-dimentional central coordinate of circle totalizer matrix of gray level image size of input, and this two-dimentional central coordinate of circle totalizer entry of a matrix element is all initialized as 0;
S2: build circle chained list; The chained list node of described round chained list comprises: central coordinate of circle (x, y), center of circle radius r and circle accumulated value a, and is empty during described round chained list initialization; Get a boundary chain C [k] in described boundary chain C [n], and 1≤k≤n, k is positive integer; If counting that this boundary chain C [k] comprises is less than threshold value T
l, then this boundary chain does not deal with, and continues to take off an edge; Otherwise be handled as follows: point step size scope of the getting [D calculating this boundary chain C [k]
k min, D
k max], D
k min, D
k maxbe positive integer; Then from D
k minstart, make variables D get [D successively
k min, D
k max] in each round values, to the respective value of each variables D, from first some P boundary chain C [k]
k1start, with the respective value of this variables D for step-length serial sampling three point in boundary chain C [k], obtain four points, wherein front three-point is for calculating central coordinate of circle (x, y) and radius r, calculate at the 4th to the distance d in the center of circle and the difference DELTA d of radius r there being the situation of solution; If | Δ d| is less than or equal to threshold value Δ D, then in circle chained list, search circle round close therewith, if find, the round accumulated value a of the correspondence in circle chained list is increased 1, otherwise inserts in circle chained list by this circle; From second some P of boundary chain C [k]
k2start with the variables D after this value for step-length repeats above process, until the point got reaches the afterbody of boundary chain C [k]; Then [D is got with variables D
k min, D
k max] next one value, and with the respective value of this variables D for step-length repeats above-mentioned process, until step-length gets D
k max;
S3: to round chained list corresponding with boundary chain C [k] in step S2, carries out ascending order arrangement to its node by center of circle radius r size; Then from first node, if the accumulated value a of circle is greater than integer thresholds M in this node, then to each element in central coordinate of circle totalizer matrix constructed in step S1, if the distance of certain marginal point in itself and boundary chain C [k] is in [r-2, r+2] in scope, and be positioned at gradient direction or the gradient negative direction of this marginal point, then 1 operation is added to the accumulator value of this element.Then take off a boundary chain and repeat step S2 and step S3, until process whole boundary chain;
S4: traversal center of circle totalizer matrix, carries out the non-maxima suppression of 4 neighborhoods, then selects accumulator value in the totalizer of the center of circle to be greater than threshold value T
mthe alternatively center of circle, the center of circle, stored in candidate center of circle chained list, and according to candidate center of circle totalizer in this candidate center of circle chained list, by accumulator value, descending sort is carried out to this candidate center of circle chained list;
S5: to the candidate center of circle chained list of the descending sort obtained in step S4, get each candidate center of circle successively, and corresponding structure radius totalizer, radius totalizer all elements is initialized as 0; Each point in edge chain C [n], calculate the distance l in itself and the candidate center of circle, and will with the immediate positive integer of l for being expressed as dl, if dl is not more than the length R of radius totalizer, then by dl totalizer in radius totalizer, namely the value of dl element adds 1, otherwise does not deal with, until process the middle institute of boundary chain C [n] a little; Then radius totalizer is traveled through, if the value of its r element is greater than threshold value T
aCCM, then think and have found a proper circle, its central coordinate of circle equals corresponding candidate's central coordinate of circle and radius equals r, by this circle stored in proper circle chained list, and is all removed by the marginal point belonging to this circle; If there is not qualified radius, then get the next candidate center of circle and repeat above process, until process all centers of circle in the chained list of the candidate center of circle.
In an embodiment of the present invention, in described step S2, getting point step size scope volume computing method is: set boundary chain to count as N
k, then D
k minbe taken as N
kthe greater in/T1 and N1, D
k maxbe taken as N
ksmaller in/T2 and N2, if D
k min>D
k max, then the value both exchanging, wherein T1, N1, T2, N2 are all the threshold values preset; And in described step S2, only retain be less than or equal to threshold value Δ D Circle Parameters to the distance d in the center of circle and the difference DELTA d of radius r at the 4th.
In an embodiment of the present invention, in step s 5, the length R of radius totalizer be original image high, wide in the greater, and constructed radius totalizer is the one dimension matrix comprising R element, its i-th element is exactly the totalizer that radius is i, wherein, and 1≤i≤R; And threshold value T
aCCMcomputing method be T
aCCM>=k*r, r are the radius in the corresponding candidate center of circle, and k is for presetting constant coefficient.
Compared to prior art, the present invention has following beneficial effect: a kind of circle detection method based on Hough transform proposed by the invention, the circle coming in detected image by utilizing Hough transform voting mechanism, point estimation radius of a circle value first by being sampled to, to determine the radius of a circle that may exist, then only these radiuses are voted according to gradient information, namely within the scope of the small neighbourhood of the radius value of gained, voting process is accelerated with gradient information, to reduce operand, efficiently reduce the impact of ballot scope and noise, proper circle is more easily detected, improve the correctness of loop truss.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of a kind of circle detection method based on Hough transform in the present invention.
Fig. 2 is the gray level image that in the present invention, input wish extracts circle.
Fig. 3 is the edge image through Canny operator extraction in the present invention.
Fig. 4 is the circle extracted in the present invention.
Embodiment
Below in conjunction with accompanying drawing, technical scheme of the present invention is specifically described.
The invention provides a kind of circle detection method based on Hough transform, it is characterized in that, as shown in Figure 1, realize as follows:
S1: after filtering and noise reduction is carried out to the gray level image of input, with the edge of this gray level image of Canny operator extraction, and the gradient vector of edge calculation point, then edge carries out the Contour extraction of 8 connections, obtain n bar 8 and be communicated with boundary chain C [n], n be more than or equal to 1 positive integer; Meanwhile, build one with the identical two-dimentional central coordinate of circle totalizer matrix of gray level image size of input, and this two-dimentional central coordinate of circle totalizer entry of a matrix element is all initialized as 0;
S2: build circle chained list; The chained list node of described round chained list comprises: central coordinate of circle (x, y), center of circle radius r and circle accumulated value a, and is empty during described round chained list initialization; Get a boundary chain C [k] in described boundary chain C [n], and 1≤k≤n, k is positive integer; If counting that this boundary chain C [k] comprises is less than threshold value T
l, then this boundary chain does not deal with, and continues to take off a boundary chain, threshold value T
lintegrality according to extracted edge is chosen, and the present embodiment gets threshold value T
l=20; Otherwise be handled as follows: point step size scope of the getting [D calculating this boundary chain C [k]
k min, D
k max], D
k min, D
k maxbe positive integer; ; Then from D
k minstart, make variables D get [D successively
k min, D
k max] in each value, to the respective value of each variables D, from first some P boundary chain C [k]
k1start, with the respective value of this variables D for step-length serial sampling three point in boundary chain C [k], obtain four points, wherein front three-point is for calculating central coordinate of circle (x, y) and radius r, calculate at the 4th to the distance d in the center of circle and the difference DELTA d of radius r there being the situation of solution; If | Δ d| is less than or equal to threshold value Δ D, then in circle chained list, search the center of circle circle round close therewith with radius, if find, the round accumulated value a of the correspondence in circle chained list is increased 1, otherwise inserted in circle chained list by this circle, and in the present embodiment, threshold value Δ D=6; From second some P of boundary chain C [k]
k2start with the variables D after this value for step-length repeats above process, until the point got reaches the afterbody of boundary chain C [k]; Then [D is got with variables D
k min, D
k max] next one value, and with the respective value of this variables D for step-length repeats above-mentioned process, until step-length gets D
k max;
S3: to round chained list corresponding with boundary chain C [k] in step S2, carries out ascending order arrangement to its node by center of circle radius r size; Then from first node, if the accumulated value a of the circle in this node is greater than integer thresholds M, then to each element in central coordinate of circle totalizer matrix constructed in step S1, if certain the marginal point distance in itself and boundary chain C [k] is in [r-2, r+2] in scope, and be positioned at gradient direction or the gradient negative direction of this marginal point, then 1 operation is added to the accumulated value of this element.In the present embodiment, threshold value M=30; Then take off a boundary chain and repeat step S2 and step S3, until process whole boundary chain; S4: traversal center of circle totalizer matrix, carries out the non-maxima suppression of 4 neighborhoods, then selects accumulator value in the totalizer of the center of circle to be greater than threshold value T
mthe center of circle alternatively the center of circle stored in candidate center of circle chained list, threshold value T
mcan choose according to practical application, in the present embodiment, get T
m=30.Then according to candidate center of circle totalizer in this candidate center of circle chained list, by accumulator value, descending sort is carried out to this candidate center of circle chained list;
S5: to the candidate center of circle chained list of the descending sort obtained in step S4, get each candidate center of circle successively, and corresponding structure radius totalizer, radius totalizer all elements is initialized as 0; Each point in edge chain C [n], calculates the distance l in itself and the candidate center of circle, and will with the immediate positive integer of l for being expressed as dl, if dl is not more than the length R of radius totalizer, then by dl totalizer in radius totalizer, namely the value of dl element adds 1, otherwise does not deal with; Until process the middle institute of boundary chain C [n] a little; Then radius totalizer is traveled through, if the value of its r element is greater than threshold value T
aCCM, then think and have found a proper circle, its central coordinate of circle equals corresponding candidate's central coordinate of circle and radius equals r, by this circle stored in proper circle chained list, and is all removed by the marginal point belonging to this circle; If there is not qualified radius, then get the next candidate center of circle and repeat above process, until process all centers of circle in the chained list of the candidate center of circle.
Further, in described step S2, getting point step size scope volume computing method is: set boundary chain to count as N
k, then D
k minbe taken as N
kthe greater in/T1 and N1, D
k maxbe taken as N
ksmaller in/T2 and N2, if D
k min>D
k max, then the value both exchanging, wherein T1, N1, T2, N2 are all the threshold values preset, and in the present embodiment, T1=20, N1=6, T2=10, N2=12; In described step S2, only retain be less than or equal to threshold value Δ D Circle Parameters to the distance d in the center of circle and the difference DELTA d of radius r at the 4th.
Further, in step s 5, the length R of radius totalizer be original image high, wide in the greater, and constructed radius totalizer is the one dimension matrix comprising R element, and its i-th element is exactly the totalizer of radius i, wherein, 1≤i≤R; And threshold value T
aCCMcomputing method be T
aCCM>=k*r, r are the radius in the corresponding candidate center of circle, and k is for presetting constant coefficient k=1.5 ~ 1.8; And in the present embodiment, concrete, T
aCCM=1.8*r.
As shown in Figure 2, be the image for carrying out loop truss, size is 307 pixel × 281 pixels, has four circles not of uniform size.Be first carry out after template size is the gaussian filtering of 9 × 9, σ 1=σ 2=2.0 to Fig. 2 as shown in Figure 3, then apply the edge image that Canny operator extraction arrives.
Be as shown in Figure 4 by the present invention propose based on Hough transform circle detection method detected by circle, 4 circles detected altogether, center of circle stain identifies, and the thick black line of the point on circle identifies.
Be more than preferred embodiment of the present invention, all changes done according to technical solution of the present invention, when the function produced does not exceed the scope of technical solution of the present invention, all belong to protection scope of the present invention.
Claims (3)
1. based on a circle detection method for Hough transform, it is characterized in that, realize as follows:
S1: after filtering and noise reduction is carried out to the gray level image of input, with the edge of this gray level image of Canny operator extraction, and the gradient vector of edge calculation point, then edge carries out the Contour extraction of 8 connections, obtain n bar 8 and be communicated with boundary chain C [n], n be more than or equal to 1 positive integer; Meanwhile, build one with the identical two-dimentional central coordinate of circle totalizer matrix of gray level image size of input, and this two-dimentional central coordinate of circle totalizer entry of a matrix element is all initialized as 0;
S2: build circle chained list; The chained list node of described round chained list comprises: central coordinate of circle (x, y), center of circle radius r and circle accumulated value a, and is empty during described round chained list initialization; Get a boundary chain C [k] in described boundary chain C [n], and 1≤k≤n, k is positive integer; If counting that this boundary chain C [k] comprises is less than threshold value T
l, then this boundary chain does not deal with, and continues to take off an edge; Otherwise be handled as follows: point step size scope of the getting [D calculating this boundary chain C [k]
k min, D
k max], D
k min, D
k maxbe positive integer; Then from D
k minstart, make variables D get [D successively
k min, D
k max] in each round values, to the respective value of each variables D, from first some P boundary chain C [k]
k1start, with the respective value of this variables D for step-length serial sampling three point in boundary chain C [k], obtain four points, wherein front three-point is for calculating central coordinate of circle (x, y) and radius r, calculate at the 4th to the distance d in the center of circle and the difference DELTA d of radius r there being the situation of solution; If | Δ d| is less than or equal to threshold value Δ D, then in circle chained list, search circle round close therewith, if find, the round accumulated value a of the correspondence in circle chained list is increased 1, otherwise inserts in circle chained list by this circle; From second some P of boundary chain C [k]
k2start with the variables D after this value for step-length repeats above process, until the point got reaches the afterbody of boundary chain C [k]; Then [D is got with variables D
k min, D
k max] next one value, and with the respective value of this variables D for step-length repeats above-mentioned process, until step-length gets D
k max;
S3: to round chained list corresponding with boundary chain C [k] in step S2, carries out ascending order arrangement to its node by center of circle radius r size; Then from first node, if the accumulated value a of circle is greater than integer thresholds M in this node, then to each element in central coordinate of circle totalizer matrix constructed in step S1, if the distance of certain marginal point in itself and boundary chain C [k] is in [r-2, r+2] in scope, and be positioned at gradient direction or the gradient negative direction of this marginal point, then 1 operation is added to the accumulator value of this element; Then take off a boundary chain and repeat step S2 and step S3, until process whole boundary chain;
S4: traversal center of circle totalizer matrix, carries out the non-maxima suppression of 4 neighborhoods, then selects accumulator value in the totalizer of the center of circle to be greater than threshold value T
mthe alternatively center of circle, the center of circle, stored in candidate center of circle chained list, and according to candidate center of circle totalizer in this candidate center of circle chained list, by accumulator value, descending sort is carried out to this candidate center of circle chained list;
S5: to the candidate center of circle chained list of the descending sort obtained in step S4, get each candidate center of circle successively, and corresponding structure radius totalizer, radius totalizer all elements is initialized as 0; Each point in edge chain C [n], calculate the distance l in itself and the candidate center of circle, and will with the immediate positive integer of l for being expressed as dl, if dl is not more than the length R of radius totalizer, then by dl totalizer in radius totalizer, namely the value of dl element adds 1, otherwise does not deal with, until process the middle institute of boundary chain C [n] a little; Then radius totalizer is traveled through, if the value of its r element is greater than threshold value T
aCCM, then think and have found a proper circle, its central coordinate of circle equals corresponding candidate's central coordinate of circle and radius equals r, by this circle stored in proper circle chained list, and is all removed by the marginal point belonging to this circle; If there is not qualified radius, then get the next candidate center of circle and repeat above process, until process all centers of circle in the chained list of the candidate center of circle.
2. a kind of circle detection method based on Hough transform according to claim 1, is characterized in that: in described step S2, gets point step size scope volume computing method to be: set boundary chain to count as N
k, then D
k minbe taken as N
kthe greater in/T1 and N1, D
k maxbe taken as N
ksmaller in/T2 and N2, if D
k min>D
k max, then the value both exchanging, wherein T1, N1, T2, N2 are all the threshold values preset; And in described step S2, only retain be less than or equal to threshold value Δ D Circle Parameters to the distance d in the center of circle and the difference DELTA d of radius r at the 4th.
3. a kind of circle detection method based on Hough transform according to claim 1, it is characterized in that: in step s 5, the length R of radius totalizer be original image high, wide in the greater, and constructed radius totalizer is the one dimension matrix comprising R element, its i-th element is exactly the totalizer that radius is i, wherein, 1≤i≤R; And threshold value T
aCCMcomputing method be T
aCCM>=k*r, r are the radius in the corresponding candidate center of circle, and k is for presetting constant coefficient.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410594409.0A CN104282027B (en) | 2014-10-29 | 2014-10-29 | Circle detecting method based on Hough transformation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410594409.0A CN104282027B (en) | 2014-10-29 | 2014-10-29 | Circle detecting method based on Hough transformation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104282027A true CN104282027A (en) | 2015-01-14 |
| CN104282027B CN104282027B (en) | 2017-05-10 |
Family
ID=52256877
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410594409.0A Active CN104282027B (en) | 2014-10-29 | 2014-10-29 | Circle detecting method based on Hough transformation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104282027B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106373117A (en) * | 2016-08-30 | 2017-02-01 | 广西科技大学 | Quick multi-circle detection method |
| CN107507185A (en) * | 2017-09-28 | 2017-12-22 | 珠海博明视觉科技有限公司 | A kind of method that quick detection suitable for industrial detection is justified |
| CN107705287A (en) * | 2017-08-24 | 2018-02-16 | 长春理工大学 | A kind of industrial products defect detection method with the Central Symmetry regularity of distribution |
| CN108053442A (en) * | 2017-10-25 | 2018-05-18 | 华南农业大学 | A kind of susceptibility piece inhibition zone size measuring method based on image |
| CN109978901A (en) * | 2019-03-07 | 2019-07-05 | 江苏亿通高科技股份有限公司 | A kind of fast, accurately circle detection and circle center locating method |
| CN110006911A (en) * | 2019-04-19 | 2019-07-12 | 天津工业大学 | It is a kind of applied to a plurality of scratch detection method in metal surface containing screw |
| CN111862037A (en) * | 2020-07-17 | 2020-10-30 | 华中科技大学无锡研究院 | Method and system for detecting geometric characteristics of precision hole type part based on machine vision |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593352A (en) * | 2009-06-12 | 2009-12-02 | 浙江大学 | Driving safety monitoring system based on face orientation and visual focus |
| US9053389B2 (en) * | 2012-12-03 | 2015-06-09 | Analog Devices, Inc. | Hough transform for circles |
| CN103971087B (en) * | 2013-07-12 | 2017-04-19 | 湖南纽思曼导航定位科技有限公司 | Method and device for searching and recognizing traffic signs in real time |
| CN103745481A (en) * | 2014-01-27 | 2014-04-23 | 哈尔滨工业大学 | Circle Hough transform high-speed VLSI (very-large-scale integration) structure |
-
2014
- 2014-10-29 CN CN201410594409.0A patent/CN104282027B/en active Active
Non-Patent Citations (2)
| Title |
|---|
| H K YUEN, ET AL.: "Comparative study of Hough Transform methods for circle finding", 《IMAGE AND VISION COMPUTING》 * |
| 尚璐 等: "改进的Hough变换圆检测算法", 《电子设计工程》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106373117A (en) * | 2016-08-30 | 2017-02-01 | 广西科技大学 | Quick multi-circle detection method |
| CN106373117B (en) * | 2016-08-30 | 2019-02-05 | 南京航空航天大学 | A kind of quickly more circle detection methods |
| CN107705287B (en) * | 2017-08-24 | 2019-10-18 | 长春理工大学 | A kind of industrial products defect detection method with the central symmetry regularity of distribution |
| CN107705287A (en) * | 2017-08-24 | 2018-02-16 | 长春理工大学 | A kind of industrial products defect detection method with the Central Symmetry regularity of distribution |
| CN107507185A (en) * | 2017-09-28 | 2017-12-22 | 珠海博明视觉科技有限公司 | A kind of method that quick detection suitable for industrial detection is justified |
| CN107507185B (en) * | 2017-09-28 | 2021-01-01 | 珠海博明视觉科技有限公司 | Method for rapidly detecting circle suitable for industrial detection |
| CN108053442A (en) * | 2017-10-25 | 2018-05-18 | 华南农业大学 | A kind of susceptibility piece inhibition zone size measuring method based on image |
| CN108053442B (en) * | 2017-10-25 | 2021-08-17 | 华南农业大学 | Image-based method for measuring size of inhibition zone of drug sensitive tablet |
| CN109978901A (en) * | 2019-03-07 | 2019-07-05 | 江苏亿通高科技股份有限公司 | A kind of fast, accurately circle detection and circle center locating method |
| CN109978901B (en) * | 2019-03-07 | 2023-06-02 | 江苏亿通高科技股份有限公司 | Quick and accurate circular detection and circle center positioning method |
| CN110006911A (en) * | 2019-04-19 | 2019-07-12 | 天津工业大学 | It is a kind of applied to a plurality of scratch detection method in metal surface containing screw |
| CN110006911B (en) * | 2019-04-19 | 2021-07-27 | 天津工业大学 | Method for detecting multiple scratches on metal surface containing screw |
| CN111862037A (en) * | 2020-07-17 | 2020-10-30 | 华中科技大学无锡研究院 | Method and system for detecting geometric characteristics of precision hole type part based on machine vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104282027B (en) | 2017-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104282027A (en) | Circle detecting method based on Hough transformation | |
| CN108960211B (en) | Multi-target human body posture detection method and system | |
| CN112837303A (en) | Defect detection method, device, equipment and medium for mold monitoring | |
| CN109559324B (en) | Target contour detection method in linear array image | |
| CN107038416B (en) | Pedestrian detection method based on binary image improved HOG characteristics | |
| CN106709500B (en) | Image feature matching method | |
| CN109919002B (en) | Yellow stop line identification method and device, computer equipment and storage medium | |
| CN111079785A (en) | Image identification method and device and terminal equipment | |
| CN104268519B (en) | Image recognition terminal and its recognition methods based on pattern match | |
| CN103745216B (en) | A kind of radar image clutter suppression method based on Spatial characteristic | |
| CN107657209B (en) | Template image registration mechanism based on finger vein image quality | |
| CN113888461A (en) | Method, system and equipment for detecting defects of hardware parts based on deep learning | |
| CN100382600C (en) | Detection method of moving object under dynamic scene | |
| Lian et al. | A novel method on moving-objects detection based on background subtraction and three frames differencing | |
| CN105184771A (en) | Adaptive moving target detection system and detection method | |
| CN103761515A (en) | Human face feature extracting method and device based on LBP | |
| CN104331885A (en) | Circular target detection method based on voting line clustering | |
| CN111950498A (en) | Lane line detection method and device based on end-to-end instance segmentation | |
| CN113378837A (en) | License plate shielding identification method and device, electronic equipment and storage medium | |
| CN112084988B (en) | Lane line instance clustering method and device, electronic equipment and storage medium | |
| CN107705414B (en) | A kind of recognition methods of bank note, device, terminal device and storage medium | |
| CN117314901A (en) | Scale-adaptive chip detection neural network system | |
| CN108241837B (en) | Method and device for detecting remnants | |
| CN113239738B (en) | Image blurring detection method and blurring detection device | |
| CN111275693B (en) | Counting method and counting device for objects in image and readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |