CN113012181A - Novel quasi-circular detection method based on Hough transformation - Google Patents
Novel quasi-circular detection method based on Hough transformation Download PDFInfo
- Publication number
- CN113012181A CN113012181A CN202110294720.3A CN202110294720A CN113012181A CN 113012181 A CN113012181 A CN 113012181A CN 202110294720 A CN202110294720 A CN 202110294720A CN 113012181 A CN113012181 A CN 113012181A
- Authority
- CN
- China
- Prior art keywords
- circle
- points
- edge
- detection
- center
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 49
- 230000009466 transformation Effects 0.000 title claims abstract description 13
- 238000003708 edge detection Methods 0.000 claims description 16
- 238000000034 method Methods 0.000 claims description 15
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 3
- 230000005764 inhibitory process Effects 0.000 claims description 3
- 238000011160 research Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 241001061260 Emmelichthys struhsakeri Species 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
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
- G06T7/13—Edge detection
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/181—Segmentation; Edge detection involving edge growing; involving edge linking
-
- 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
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
Abstract
The invention discloses a novel circle-like shape detection method based on Hough transformation, which combines Hough transformation with a circle-like shape detection algorithm based on a circumferential angle and a circle-like shape detection algorithm based on a circle radius respectively to judge the circle-like shape in a one-dimensional image, and has the advantages of high operation speed, small occupied memory, higher detection precision under the condition of higher image definition, higher practical value and stronger universal applicability.
Description
Technical Field
The invention relates to the field of image recognition, in particular to a novel circle-like detection method based on Hough transformation.
Background
At present, many targets need to perform rapid and accurate circle-like detection, so that the rapid and accurate circle-like detection from one picture is a necessary research problem, but under the condition that the data volume of the picture is increasing, all the existing circle or ellipse detection methods have some problems in different aspects.
In various researches, the commonly used algorithms are a classical hough algorithm, a random hough algorithm and a generalized hough algorithm. The learners use the characteristics of the closed boundary of the ellipse to provide a method for efficiently detecting the ellipse, the concept of an ellipse boundary chain code is provided to carry out edge detection and tracking on the ellipse, the edge detection is clearer by using filtering to eliminate noise and non-closed data so as to form an ellipse boundary chain code, the even points are searched on the chain code and Hough transformation is carried out, and therefore the ellipse is accurately detected. However, the algorithm has time complexity of O (n × n) and is slow in operation. Starting from the arc length of the ellipse, the general equation and parameters of the ellipse are determined by the position information and the gradient information of the edge point of the same arc length, and the effectiveness and the detection time of the algorithm are superior to those of the general detection algorithm. But the recognition accuracy of the algorithm is low for non-standard ellipses.
The method combines the specific characteristics of the similar-circle shape, applies a geometric concept, performs gray level conversion and binarization on the image to be identified, sequentially obtains the edge of the image, discards more dispersed points to form the edge of the image, and extracts the similar-circle parameters by using Hough transform to obtain the center of the similar-circle.
Disclosure of Invention
In order to overcome the defects in the prior art, the operation speed is low, and the detection success rate for similar circles is low, the invention provides a novel similar circle detection method based on Hough transformation, and the method is applied to the detection of meteor craters on the lunar surface so as to meet the obstacle avoidance requirement of a lunar rover.
In order to achieve the above purpose, the technical solution for solving the technical problem is as follows:
the invention discloses a novel circle-like detection method based on Hough transformation, which is combined with a circle-like detection method based on a circumferential angle, and comprises the following steps:
step 1: acquiring the circle center of a quasi-circle, wherein the acquired image boundary points need to be filtered to acquire a continuous boundary, assuming the circle center of the quasi-circle as an O point, establishing a horizontal scanning line on a plane where the image is located, sequentially scanning downwards from the edge of the quasi-circle, counting the middle points in a one-dimensional space by adopting a Hough conversion method, and taking the middle point corresponding to the maximum counting value as the circle center 0;
step 2: obtaining the radius of a circle-like, finding out the outline of the target graph through edge detection, sequentially traversing each point on the outline, carrying out numerical comparison to obtain two points at the farthest distance on the outline, taking any point on the circle except the two points, and regarding the connecting line of the two points at the farthest distance as the diameter of the circle-like;
and step 3: and judging whether the angle range of the diameter is about 80-100 degrees, if so, drawing the edge and giving a central point O.
Further, the edge detection step in step 2 is as follows:
step 21: using Gaussian filtering to remove noise to achieve the purpose of smoothing the image;
step 22: searching the intensity gradient of the quasi-circular boundary, namely dividing the image according to different pixel values;
step 23: tracking edge points along the edge direction by using a non-maximum inhibition technology to eliminate edge false detection;
step 24: a double threshold method is used to determine the possible boundaries and to map the boundaries.
The invention also discloses a novel similar circle detection method based on Hough transformation, which is combined with a similar circle detection method based on circle radius, and comprises the following steps:
step A: acquiring the center of a circle-like, wherein boundary points of the acquired image need to be filtered to acquire a continuous boundary, assuming the center of the circle-like as an O point, a horizontal scanning line is established on a plane where the image is located, the horizontal scanning line is sequentially scanned downwards from the edge of the circle-like, in a one-dimensional space, the midpoint counting adopts a Hough conversion method, and the midpoint corresponding to the maximum counting value can be regarded as the center of the circle-like O;
and B: acquiring the radius of a similar circle, identifying the image contour through edge detection, sequentially traversing points on the contour to obtain two points with the farthest distance on the contour, setting the two points as AB, and assuming a midpoint O of the AB as a circle center, wherein the distance between the two points can be assumed as a diameter;
and C: if the distance from each point to the midpoint on the identified edge fluctuates around the radius value and the radius data is normalized to solve the problem of inconsistent fluctuation ranges of large diameter and small diameter, the graph-like circle can be judged.
Due to the adoption of the technical scheme, compared with the prior art, the invention has the following advantages and positive effects:
1. the invention provides a quasi-circular detection method for detecting according to a quasi-circular contour aiming at the problems of instability, high time and space consumption and the like of Hough transform in circle detection.
2. The Hough transform is combined with other two circle-like detection methods (a circle-like detection method based on a circumferential angle and a circle-like detection method based on a circle radius), and the Hough transform is applied to the detection of meteor craters to judge the circle-like in the one-dimensional image, so that the Hough transform-based meteor crater detection method is high in operation speed, small in occupied memory, high in detection precision under the condition of high image definition, high in practical value and high in general applicability. The method has good identification effect on complex lunar surface terrain, and lays an important theoretical foundation for obstacle avoidance of a lunar vehicle later.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort. In the drawings:
FIG. 1 is a detailed block diagram of the detection method of the present invention;
FIG. 2 is a simulation diagram for determining a circumferential angle according to the present invention;
FIG. 3 is a simulation diagram of the determination of circle radius in the present invention;
FIG. 4 is a meteorite crater artwork in the present invention;
FIG. 5 is a meteorite crater detection map of the present invention;
FIG. 6 is a graph of the results of a conventional algorithm;
fig. 7 is a graph of the results of the algorithm operation of the present invention.
Detailed Description
While the embodiments of the present invention will be described and illustrated in detail with reference to the accompanying drawings, it is to be understood that the invention is not limited to the specific embodiments disclosed, but is intended to cover various modifications, equivalents, and alternatives falling within the scope of the invention as defined by the appended claims.
Example one
As shown in figure 1, the invention discloses a novel circle-like detection method based on Hough transformation, which is combined with a circle-like detection method based on a circumferential angle, and comprises the following steps:
step 1: acquiring the center of a circle-like, wherein boundary points of the acquired image need to be filtered to acquire a continuous boundary, assuming the center of the circle-like as an O point, a horizontal scanning line is established on a plane where the image is located, the horizontal scanning line is sequentially scanned downwards from the edge of the circle-like, in a one-dimensional space, the midpoint counting adopts a Hough conversion method, and the midpoint corresponding to the maximum counting value can be regarded as the center of the circle-like O;
step 2: obtaining the radius of a circle-like, finding out the outline of the target graph through edge detection, sequentially traversing each point on the outline, carrying out numerical comparison to obtain two points at the farthest distance on the outline, taking any point on the circle except the two points, and regarding the connecting line of the two points at the farthest distance as the diameter of the circle-like;
and step 3: and judging whether the angle range corresponding to the diameter is about 80-100 degrees, if so, drawing the edge and giving a central point O, wherein the specific parameters are shown in figure 2.
In the detection method, edge detection needs to be performed on an original image, and under the condition that original image attributes are kept, the data scale of the image is reduced remarkably, and the specific edge detection steps are as follows:
step 21: using Gaussian filtering to remove noise to achieve the purpose of smoothing the image;
step 22: searching the intensity gradient of the quasi-circular boundary, namely dividing the image according to different pixel values;
step 23: tracking edge points along the edge direction by using a non-maximum inhibition technology to eliminate edge false detection;
step 24: a double threshold method is used to determine the possible boundaries and to map the boundaries.
After the edge detection is successful, the contour of the point to be determined in the picture is searched, each point on the contour is sequentially traversed to obtain two farthest points on the contour, such as AB shown in fig. 2, any point (except the two points) is taken on the circle, whether the angle a corresponding to the edge of AB is between 80 degrees and 100 degrees is judged, and if the angle is in the range, the graph can be marked and can be used for further identification.
Example two
Continuing to refer to fig. 1, the invention also discloses a novel circle-like detection method based on Hough transformation, which is combined with a circle-like detection method based on circle radius, wherein any line segment from the center to the periphery of the circle-like detection method is equal based on the characteristics of the circle radius, and the figure can be approximately regarded as a circle if any point on the figure to the determined midpoint fluctuates in the radius, and the method specifically comprises the following steps:
step A: acquiring the center of a circle-like, wherein boundary points of the acquired image need to be filtered to acquire a continuous boundary, assuming the center of the circle-like as an O point, a horizontal scanning line is established on a plane where the image is located, the horizontal scanning line is sequentially scanned downwards from the edge of the circle-like, in a one-dimensional space, the midpoint counting adopts a Hough conversion method, and the midpoint corresponding to the maximum counting value can be regarded as the center of the circle-like O;
and B: acquiring the radius of a similar circle, identifying the image contour through edge detection, sequentially traversing points on the contour to obtain two points with the farthest distance on the contour, and assuming that the midpoint O of the AB is the center of the circle and the distance between the two points can be assumed to be the diameter as shown in AB in FIG. 3;
and C: if the distance from each point to the midpoint on the identified edge fluctuates around the radius value, that is, the lengths of m and n are approximately equal, normalization processing needs to be performed on the radius data to solve the problem of inconsistent fluctuation ranges of large diameter and small diameter, and then the graph can be judged to be similar to a circle, and specific parameters are shown in fig. 3.
The implementation of the invention is completed in three stages, as follows:
the first stage is as follows: specific implementation of the algorithm
Firstly, edge detection is required to be carried out on an original image, and the edge of the similar circle is determined after binarization operation. For any non-zero point of the image edge, a Sobel operator is required, and for each pixel specified by the parameters, the pixels are overlapped in an accumulator and marked. The inverse resolution of the accumulator image is 1, the high threshold of the Canny edge function is set to 60, and the circle center detection threshold is set to 26. The circle center detection threshold needs to be set according to the size of a circle in the image, and when the circle in the picture is smaller, the value should be set smaller, so as to avoid the generation of noise.
The circumference angle algorithm is realized by the following steps:
1. finding out the outline of the target graph through edge detection;
2. sequentially traversing each point on the contour;
3. carrying out numerical comparison to obtain two points with the farthest distance on the contour;
4. taking any point on the circle (except the two points), and regarding the connecting line of the two points at the farthest distance as the diameter of the circle;
5. and judging whether the angle range of the diameter is about 80-100 degrees or not.
The circle radius algorithm is realized by the following steps:
1. identifying the image contour through edge detection;
2. sequentially traversing points on the contour;
3. judging to obtain the diameters of two points which are farthest away on the contour;
4. it is determined whether the distance of each point on the identified edge to the midpoint fluctuates about the radius value.
And a second stage: specific application of algorithm
The method for identifying and detecting the similar circle is widely applied to industrial detection lines, biomedical monitoring equipment and automatic assembly lines.
The problem that the relevant research of obstacle avoidance of the lunar rover cannot be ignored is the research on complex terrains, and the invention applies two similar circle detection algorithms to the identification of the lunar surface meteor crater and obtains good effect. In the aspect of route planning of a lunar vehicle, identification and detection of meteor craters can provide better data basis for route planning to avoid obstacles, and the specific application is shown in fig. 4 and 5.
And a third stage: accuracy analysis of circle-like recognition
The recognition accuracy of the invention for the quasi-circular shape is shown by the specific implementation chart, and according to fig. 6 and 7, the accuracy of the traditional algorithm for the quasi-circular shape detection is lower than that of the algorithm proposed by the invention.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (3)
1. A novel circle-like detection method based on Hough transformation is combined with a circle-like detection method based on circumferential angles, and is characterized by comprising the following steps:
step 1: acquiring the center of a circle-like, wherein boundary points of the acquired image need to be filtered to acquire a continuous boundary, assuming the center of the circle-like as an O point, a horizontal scanning line is established on a plane where the image is located, the horizontal scanning line is sequentially scanned downwards from the edge of the circle-like, in a one-dimensional space, the midpoint counting adopts a Hough conversion method, and the midpoint corresponding to the maximum counting value can be regarded as the center of the circle-like O;
step 2: obtaining the radius of a circle-like, finding out the outline of the target graph through edge detection, sequentially traversing each point on the outline, carrying out numerical comparison to obtain two points at the farthest distance on the outline, taking any point on the circle except the two points, and regarding the connecting line of the two points at the farthest distance as the diameter of the circle-like;
and step 3: and judging whether the angle range of the diameter is about 80-100 degrees, if so, drawing the edge and giving a central point O.
2. The novel Hough transform-based quasi-circular detection method according to claim 1, wherein the edge detection step in step 2 is as follows:
step 21: using Gaussian filtering to remove noise to achieve the purpose of smoothing the image;
step 22: searching the intensity gradient of the quasi-circular boundary, namely dividing the image according to different pixel values;
step 23: tracking edge points along the edge direction by using a non-maximum inhibition technology to eliminate edge false detection;
step 24: a double threshold method is used to determine the possible boundaries and to map the boundaries.
3. A novel round-like detection method based on Hough transformation is combined with a round-like detection method based on circle radius, and is characterized by comprising the following steps:
step A: acquiring the center of a circle-like, wherein boundary points of the acquired image need to be filtered to acquire a continuous boundary, assuming the center of the circle-like as an O point, a horizontal scanning line is established on a plane where the image is located, the horizontal scanning line is sequentially scanned downwards from the edge of the circle-like, in a one-dimensional space, the midpoint counting adopts a Hough conversion method, and the midpoint corresponding to the maximum counting value can be regarded as the center of the circle-like O;
and B: acquiring the radius of a similar circle, identifying the image contour through edge detection, sequentially traversing points on the contour to obtain two points with the farthest distance on the contour, setting the two points as AB, and assuming a midpoint O of the AB as a circle center, wherein the distance between the two points can be assumed as a diameter;
and C: if the distance from each point to the midpoint on the identified edge fluctuates around the radius value and the radius data is normalized to solve the problem of inconsistent fluctuation ranges of large diameter and small diameter, the graph-like circle can be judged.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110294720.3A CN113012181B (en) | 2021-03-18 | 2021-03-18 | Novel quasi-circular detection method based on Hough transformation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110294720.3A CN113012181B (en) | 2021-03-18 | 2021-03-18 | Novel quasi-circular detection method based on Hough transformation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113012181A true CN113012181A (en) | 2021-06-22 |
CN113012181B CN113012181B (en) | 2023-09-19 |
Family
ID=76402946
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110294720.3A Active CN113012181B (en) | 2021-03-18 | 2021-03-18 | Novel quasi-circular detection method based on Hough transformation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113012181B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113570593A (en) * | 2021-08-10 | 2021-10-29 | 深圳诺博医疗科技有限公司 | Accurate medicament counting method and device, computer equipment and storage medium |
CN114359320A (en) * | 2021-12-15 | 2022-04-15 | 哈尔滨工业大学 | Moon detector robust ring mountain detection method and aircraft navigation method |
CN113570593B (en) * | 2021-08-10 | 2024-05-14 | 深圳诺博医疗科技有限公司 | Accurate counting method and device for medicament, computer equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5189711A (en) * | 1989-11-24 | 1993-02-23 | Isaac Weiss | Automatic detection of elliptical shapes |
DE102015205505A1 (en) * | 2015-03-26 | 2016-09-29 | Mando Corporation | Method and apparatus for detecting elliptical structures in an image |
CN108986038A (en) * | 2018-06-01 | 2018-12-11 | 浙江大学 | A kind of wheel-hub contour detection method based on Improved Hough Transform |
CN110060259A (en) * | 2019-04-04 | 2019-07-26 | 东华大学 | A kind of fish eye lens effective coverage extracting method based on Hough transformation |
-
2021
- 2021-03-18 CN CN202110294720.3A patent/CN113012181B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5189711A (en) * | 1989-11-24 | 1993-02-23 | Isaac Weiss | Automatic detection of elliptical shapes |
DE102015205505A1 (en) * | 2015-03-26 | 2016-09-29 | Mando Corporation | Method and apparatus for detecting elliptical structures in an image |
CN108986038A (en) * | 2018-06-01 | 2018-12-11 | 浙江大学 | A kind of wheel-hub contour detection method based on Improved Hough Transform |
CN110060259A (en) * | 2019-04-04 | 2019-07-26 | 东华大学 | A kind of fish eye lens effective coverage extracting method based on Hough transformation |
Non-Patent Citations (2)
Title |
---|
屈汉章;王洋;吴成茂;: "基于直径累积的霍夫变换检测圆算法", 西安邮电大学学报, no. 05 * |
秦开怀;王海颍;郑辑涛;: "一种基于Hough变换的圆和矩形的快速检测方法", 中国图象图形学报, no. 01 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113570593A (en) * | 2021-08-10 | 2021-10-29 | 深圳诺博医疗科技有限公司 | Accurate medicament counting method and device, computer equipment and storage medium |
CN113570593B (en) * | 2021-08-10 | 2024-05-14 | 深圳诺博医疗科技有限公司 | Accurate counting method and device for medicament, computer equipment and storage medium |
CN114359320A (en) * | 2021-12-15 | 2022-04-15 | 哈尔滨工业大学 | Moon detector robust ring mountain detection method and aircraft navigation method |
CN114359320B (en) * | 2021-12-15 | 2023-02-03 | 哈尔滨工业大学 | Moon detector robust ring mountain detection method and aircraft navigation method |
Also Published As
Publication number | Publication date |
---|---|
CN113012181B (en) | 2023-09-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108921176B (en) | Pointer instrument positioning and identifying method based on machine vision | |
Kong et al. | Generalizing Laplacian of Gaussian filters for vanishing-point detection | |
CN107292310B (en) | Visual positioning and automatic reading method for circular pointer type dial plate | |
CN106407924A (en) | Binocular road identifying and detecting method based on pavement characteristics | |
CN107748855B (en) | Detection method for two-dimensional code image finding graph | |
CN108447016B (en) | Optical image and SAR image matching method based on straight line intersection point | |
CN107609510B (en) | Positioning method and device for lower set of quayside container crane | |
CN107957246B (en) | binocular vision-based method for measuring geometric dimension of object on conveyor belt | |
CN107192716A (en) | A kind of workpiece, defect quick determination method based on contour feature | |
CN111080661A (en) | Image-based line detection method and device and electronic equipment | |
CN112598922B (en) | Parking space detection method, device, equipment and storage medium | |
CN109300104B (en) | Angular point detection method and device | |
CN114240845B (en) | Light cutting method surface roughness measurement method applied to cutting workpiece | |
CN111652790B (en) | Sub-pixel image registration method | |
CN107564006B (en) | Circular target detection method utilizing Hough transformation | |
CN106778551A (en) | A kind of fastlink and urban road Lane detection method | |
CN111582000B (en) | Bar code positioning method, bar code positioning device and related equipment | |
CN113012181A (en) | Novel quasi-circular detection method based on Hough transformation | |
CN111489389A (en) | Light spot center detection method | |
US9830532B1 (en) | Object detection in images using distance maps | |
CN109544513A (en) | A kind of steel pipe end surface defect extraction knowledge method for distinguishing | |
CN107710229B (en) | Method, device and equipment for recognizing shape in image and computer storage medium | |
CN112991327B (en) | Steel grid welding system, method and terminal equipment based on machine vision | |
CN112767425A (en) | Parking space detection method and device based on vision | |
CN113298725A (en) | Correction method for superposition error of ship icon image |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |