CN108492306A - A kind of X-type Angular Point Extracting Method based on image outline - Google Patents
A kind of X-type Angular Point Extracting Method based on image outline Download PDFInfo
- Publication number
- CN108492306A CN108492306A CN201810187819.1A CN201810187819A CN108492306A CN 108492306 A CN108492306 A CN 108492306A CN 201810187819 A CN201810187819 A CN 201810187819A CN 108492306 A CN108492306 A CN 108492306A
- Authority
- CN
- China
- Prior art keywords
- point
- image
- profile
- angle point
- angle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The present invention proposes a kind of X-type Angular Point Extracting Method based on image outline, it is related to technical field of vision detection, the present invention traverses Harris angle points successively by building window on certain profile, when it is 7 to traverse angle point number, it is X-type angle point to take the point nearest apart from 7 angular coordinate mean values, and traversal next profile completes the extraction of all profile X-type angle points when extraction angle point number is 4；The present invention can filter the pseudo- angle point in image, the angle point that can be accurately detected in image, and accuracy of detection reaches 0.09 pixel；By the periphery background for setting angle point so that characteristic point is not influenced mark point in by environment, and angle point is detected in complicated scene；It is precisely extracted in conjunction with Harris and the advantages of locations of contours, sub-pixel detection is carried out to the angle point of laying, improves the coordinate precision of angle point；The threshold value that image preprocessing is adjusted by dynamic, reduces the influence of illumination in different scenes, improves the success rate of Corner Detection.
Description
Technical field
The present invention relates to technical field of vision detection, and in particular to a kind of X-type Angular Point Extracting Method based on image outline.
Background technology
With the continuous development of science and technology, video camera has also obtained development at full speed, the type and function of video camera
Also more and more, people have also enjoyed the various facilities that development in science and technology is brought.For example, people can utilize video camera setting
On vehicle, the environmental information in parking lot is known by way of camera acquisition image.
Current video camera needs to demarcate parameter when in use, to obtain more accurate parameter, and parameter meter
Calculation is directly influenced by angle point accuracy, and this requires the accuracy of angle point grid is higher.Traditional angle point grid technology is based on
The angular-point detection method of section test, but image scene is complicated, in addition the template size of section test, gray scale difference threshold values are difficult to select
Etc. reasons, cause during detecting X-type angle point, be easy to cause missing inspection, more inspection, more pseudo- angle point occur.
Invention content
In view of the deficiencies of the prior art, the present invention proposes a kind of X-type Angular Point Extracting Method based on image outline, Neng Goujing
Really X-type angle point in identification image, guarantee is provided for video camera Accurate Calibration.
A kind of X-type Angular Point Extracting Method based on image outline, includes the following steps：
Step 1 obtains pending image, using Harris Angular Point Extracting Methods to Harris angle points all in image into
Row extraction；
Step 2, setting threshold value, to image carry out binary conversion treatment, obtain black white image, to black white image carry out corrosion and
Opening operation processing；
All profiles in step 3, extraction black white image, by way of building window, on a certain profile successively time
Go through all Harris angle points；
Step 4 judges to traverse whether obtained Harris angle points number is 7, if so, taking equal apart from 7 angular coordinates
It is X-type angle point to be worth nearest point, and traverses next profile and otherwise traverse next profile；
Step 5, after having traversed all profiles, whether the number for judging to extract the X-type angle point is 4, if so, complete institute
There is the extraction of profile X-type angle point, otherwise, dynamic adjusts the threshold value and the window length of side of binary conversion treatment setting, until extraction X-type angle
The number of point is 4, that is, the 4 X-type angle points marked all detect, detection finishes.
Described in step 3 extraction black white image in all profiles, by way of building window, on a certain profile according to
Secondary all Harris angle points of traversal；
Specially：
It is arbitrarily taken on profile a bit, the point centered on the point, the length of side is set, build window；
Using central point as starting point, the pixel along profile cycling among windows is played；
If traversing out, there are Harris angle points to fall into window, records, and deletes the Harris angle points in the picture, after
The continuous pixel along profile cycling among windows, until being back to starting point.
Described in step 4 otherwise, next profile is traversed, further includes the Harris angle points to undelete.
Advantage of the present invention：
The present invention proposes a kind of X-type Angular Point Extracting Method based on image outline, can filter the pseudo- angle point in image, energy
The angle point being enough accurately detected in image, accuracy of detection can reach 0.09 pixel；It is (black by the periphery background for setting angle point
White colour) so that characteristic point is not influenced mark point in by environment substantially, can detect angle point in complicated scene；Knot
The advantages of closing the accurate extractions of Harris and locations of contours can carry out sub-pixel detection to the angle point of laying, improve angle point
Coordinate precision；The threshold value that image preprocessing is adjusted by dynamic, reduces the influence of illumination in different scenes, improves angle point
The success rate of detection.
Description of the drawings
Fig. 1 is the X-type Angular Point Extracting Method flow chart based on image outline of an embodiment of the present invention；
Fig. 2 is the pending image schematic diagram of an embodiment of the present invention；
Fig. 3 is image schematic diagram after the Harris Corner Detections of an embodiment of the present invention；
Fig. 4 is that schematic diagram is interrupted in the X-type angle point region of an embodiment of the present invention；
The corrosion and the X-type angle point area schematic after opening operation that Fig. 5 is an embodiment of the present invention；
Fig. 6 is the first-order difference template schematic diagram of an embodiment of the present invention, wherein figure (a) is the first-order difference of X-direction
Template schematic diagram, figure (b) are the first-order difference template schematic diagram of Y-direction；
Fig. 7 is the direction schematic diagram of the gradient of an embodiment of the present invention, wherein figure (a) is the sector point of gradient direction
Schematic diagram is cut, figure (b) is the pixel selection schematic diagram of gradient direction；
Fig. 8 is the contours extract schematic diagram of an embodiment of the present invention；
Fig. 9 is that schematic diagram is completed in the extraction of the profile X-type angle point of an embodiment of the present invention.
Specific implementation mode
An embodiment of the present invention is described further below in conjunction with the accompanying drawings.
In the embodiment of the present invention, the X-type Angular Point Extracting Method based on image outline, as shown in Figure 1, including the following steps：
Step 1 obtains pending image, using Harris Angular Point Extracting Methods to Harris angle points all in image into
Row extraction；
In present example, it is illustrated in figure 2 pending image, the coordinate bit of X-type angle point is determined by Harris operators
It sets, is as follows：
Coordinate system is established by origin of any point in image, it preferably can be with any one angle at four angles in image
For origin, wherein the horizontal direction of image is X-axis, and the vertical direction of image is Y-axis, and angle point inspection is carried out by Harris operators
It surveys., can be by the angle point in Harris operator detection images after establishing coordinate system, specific detection process is as follows：
The direction x, y difference is calculated first with gradient operator, the gamma function of each pixel in image is recycled to calculate
The x-axis of each pixel in image, y-axis direction gradient Ix、Iy, calculation formula is as follows：
Each pixel is calculated separately in image again in the gradient product I of both directionx 2、Iy 2And Ixy, calculation formula is such as
Under：
Ix 2=Ix*Ix(3）
Iy 2=Iy *Iy(4）
Ixy=Ix *Iy (5)
It can be obtained by 3 width new images in this way, the corresponding attribute of each pixel is respectively each pixel two sides
To gradient product Ix 2、Iy 2And Ixy, gaussian filtering is then carried out respectively to 3 width images, is specifically as follows：
Gaussian kernel is generated respectively to I using Gaussian functionx 2、Iy 2And IxyFiltering, wherein Gaussian function formula are：
The angle point amount R of each pixel can be calculated according to formula (6)~formula (9), calculation formula is as follows：
In present example, appropriately sized window (such as 5*5) is selected, all pixels in image is traversed, calculates the window
The local maximum of interior pixel corner amount, if this value is more than some threshold values thresh, (specific value is in present example
0.05), then it is assumed that the pixel is a candidate point；It sorts according to the angle point amount size of candidate point, takes interested preceding N (this hair
Specific value is 200) a point as last Harris angle points in bright example；In present example, it is illustrated in figure 3 Harris
Image after Corner Detection；
Step 2, setting threshold value (specific value is 50 in present example), carry out binary conversion treatment to image, obtain black
White image carries out corrosion to black white image and opening operation is handled；
In the embodiment of the present invention, for preferably detection edge, gray level image to be detected is subjected to binary conversion treatment；Carry out
During binary conversion treatment, from the image-region of video camera farther out since visual angle is smaller, inclined degree is excessive, lead to X-type angle
Point black region interruption, this will produce the X-type region contour when extracting profile and is not connected to, and eventually lead to outline and angle point inspection
Dendrometry loses.In order to solve this problem, corrosion is carried out to the image after binary conversion treatment and opening operation operates.
In present example, it is illustrated in figure 4 X-type angle point region interruption schematic diagram.It is illustrated in figure 5 corrosion and opening operation
X-type angle point area schematic afterwards.
All profiles in step 3, extraction black white image, by way of building window, on a certain profile successively time
Go through all Harris angle points；Specially：It is arbitrarily taken on profile a bit, the point centered on the point, the length of side is set, build window；
Using central point as starting point, the pixel along profile cycling among windows is played；If traversing out, there are Harris angle points to fall into window,
Record, and the Harris angle points are deleted in the picture, continue the pixel along profile cycling among windows, until being back to starting point；
In present example, after carrying out etching operation to image to be detected, image side is obtained by canny operators
Edge uses Gaussian filter function smoothed image first, then increases using as schemed first-order difference template shown in (a) and figure (b) in Fig. 6
Strong image border calculates amplitude and the direction of gradient, and non-pole is carried out to the gradient of calculating using the direction of gradient as shown in Figure 7
Big value inhibits, and retains the maximum pixel of partial gradient.
In present example, four kinds of combinations of figure (b) template in 4 pairs of sectors difference corresponding diagram 7 of (a) are schemed in Fig. 7；It utilizes
Gradient direction selects the label of sector, and the gradient width of the gradient magnitude and neighbor pixel is then compared according to the direction of label
Value retains initial value, the value is otherwise set to 0 if the value is both bigger than other.If for example, the gradient side of template intermediate point
To sector label 3 is directed toward, then compare the gradient magnitude of 4,8 points of intermediate point and neighborhood, if middle point value is both greater than 4,8 point values, M
It is constant, otherwise M=0.
In present example, gradient magnitude is handled by bilinearity threshold values, is specifically as follows：Selection upper limit value is under respectively
Limit value, if the gradient of a pixel is more than upper limit threshold values, then it is assumed that the point is edge pixel point；If being less than lower limit threshold values,
Then the point is abandoned；If the value is more than lower limit threshold values and is less than upper limit threshold values, only when the point and higher than upper limit threshold values
Just received when pixel is connected, the straight line where the pixel being connected can be determined as image border.
In present example, profile can be extracted after obtaining edge image to image by leading to, to obtain X-type angle point area
At least one region contour of domain black and white lattice, as shown in Figure 8；It, can be right after getting at least one X-type angle point region contour
Accessed region contour is matched, to obtain matching area profile；It is specifically as follows：Successively according to the preset window length of side
Each X-type angle point region contour at least one X-type angle point region contour is traversed, judges to whether there is in window appli
Angle point；If there are angle point, the region contour comprising 7 initial angle points is determined as matching area profile；
In present example, it is logical from Fig. 3 and Fig. 4 as can be seen that there is 7 angles on each X-type angle point region contour
Point.That is, having found the point demarcated in image as long as matching the profile with 7 initial angle points.When matching profile,
The suitable window length of side (specific value is 3 in present example) can be selected, traverses all pixels point of profile successively, is found
Whether there is angle point in the neighborhood of pixel points.In specific implementation process, due to after etching operation, X-type angle point region contour
As the one of connection, this profile for allowing for detecting becomes larger with a distance from real edges, so needing rational selection window
The length of side detects angle point.If there are angle point, add up the number of angle point, and each angle point only counts once, until having traversed
Some pixels.
Step 4 judges to traverse whether obtained Harris angle points number is 7, if so, taking equal apart from 7 angular coordinates
It is X-type angle point to be worth nearest point, and traverses next profile, and otherwise, the Harris angle points to undelete traverse next wheel
It is wide；
Step 5, after having traversed all profiles, whether the number for judging to extract the X-type angle point is 4, if so, complete institute
There is the extraction of profile X-type angle point, as shown in figure 9, otherwise, dynamic adjusts the threshold value and the window length of side of binary conversion treatment setting, directly
Number to extraction X-type angle point is 4, that is, the 4 X-type angle points marked all detect that detection finishes.
Claims (3)
1. a kind of X-type Angular Point Extracting Method based on image outline, which is characterized in that include the following steps：
Step 1 obtains pending image, is carried to Harris angle points all in image using Harris Angular Point Extracting Methods
It takes；
Step 2, setting threshold value, carry out binary conversion treatment to image, obtain black white image, corroded and opened fortune to black white image
Calculation is handled；
All profiles in step 3, extraction black white image traverse institute successively by way of building window on a certain profile
There are Harris angle points；
Step 4 judges to traverse whether obtained Harris angle points number is 7, if so, taking apart from 7 angular coordinate mean values most
Close point is X-type angle point, and traverses next profile and otherwise traverse next profile；
Step 5, after having traversed all profiles, whether the number for judging to extract the X-type angle point is 4, if so, completing all wheels
The extraction of wide X-type angle point, otherwise, dynamic adjust the threshold value and the window length of side of binary conversion treatment setting, until extraction X-type angle point
Number is 4, that is, 4 marked X-type angle point all detects that detection finishes.
2. the X-type Angular Point Extracting Method according to claim 1 based on image outline, which is characterized in that described in step 3
All profiles in extraction black white image traverse all angles Harris successively by way of building window on a certain profile
Point；
Specially：
It is arbitrarily taken on profile a bit, the point centered on the point, the length of side is set, build window；
Using central point as starting point, the pixel along profile cycling among windows is played；
If traversing out, there are Harris angle points to fall into window, records, and deletes the Harris angle points in the picture, continues edge
Pixel in profile cycling among windows, until being back to starting point.
3. the X-type Angular Point Extracting Method according to claim 1 based on image outline, which is characterized in that described in step 4
Otherwise, next profile is traversed, further includes the Harris angle points to undelete.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810187819.1A CN108492306A (en) | 2018-03-07 | 2018-03-07 | A kind of X-type Angular Point Extracting Method based on image outline |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810187819.1A CN108492306A (en) | 2018-03-07 | 2018-03-07 | A kind of X-type Angular Point Extracting Method based on image outline |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108492306A true CN108492306A (en) | 2018-09-04 |
Family
ID=63341848
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810187819.1A Pending CN108492306A (en) | 2018-03-07 | 2018-03-07 | A kind of X-type Angular Point Extracting Method based on image outline |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108492306A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109829910A (en) * | 2019-02-01 | 2019-05-31 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of PCB defect inspection method searched based on neighborhood |
| CN111429450A (en) * | 2020-04-10 | 2020-07-17 | 展讯通信（上海）有限公司 | Corner point detection method, system, equipment and storage medium |
| CN112387982A (en) * | 2020-10-21 | 2021-02-23 | 上海交通大学 | Laser additive process power combined regulation and control method |
| CN113310987A (en) * | 2020-02-26 | 2021-08-27 | 保定市天河电子技术有限公司 | Tunnel lining surface detection system and method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101887586A (en) * | 2010-07-30 | 2010-11-17 | 上海交通大学 | Self-adaptive angular-point detection method based on image contour sharpness |
| CN102169578A (en) * | 2011-03-16 | 2011-08-31 | 内蒙古科技大学 | Non-rigid medical image registration method based on finite element model |
| CN103345755A (en) * | 2013-07-11 | 2013-10-09 | 北京理工大学 | Chessboard angular point sub-pixel extraction method based on Harris operator |
| CN104075659A (en) * | 2014-06-24 | 2014-10-01 | 华南理工大学 | Three-dimensional imaging recognition method based on RGB structure light source |
| CN105139412A (en) * | 2015-09-25 | 2015-12-09 | 深圳大学 | Hyperspectral image corner detection method and system |
| CN106599483A (en) * | 2016-12-16 | 2017-04-26 | 鞍钢集团矿业有限公司 | Processing method of open-pit mine slope monitoring plane data based on measurement robot |
| CN107369162A (en) * | 2017-07-21 | 2017-11-21 | 华北电力大学（保定） | A kind of generation method and system of insulator candidate target region |
-
2018
- 2018-03-07 CN CN201810187819.1A patent/CN108492306A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101887586A (en) * | 2010-07-30 | 2010-11-17 | 上海交通大学 | Self-adaptive angular-point detection method based on image contour sharpness |
| CN102169578A (en) * | 2011-03-16 | 2011-08-31 | 内蒙古科技大学 | Non-rigid medical image registration method based on finite element model |
| CN103345755A (en) * | 2013-07-11 | 2013-10-09 | 北京理工大学 | Chessboard angular point sub-pixel extraction method based on Harris operator |
| CN104075659A (en) * | 2014-06-24 | 2014-10-01 | 华南理工大学 | Three-dimensional imaging recognition method based on RGB structure light source |
| CN105139412A (en) * | 2015-09-25 | 2015-12-09 | 深圳大学 | Hyperspectral image corner detection method and system |
| CN106599483A (en) * | 2016-12-16 | 2017-04-26 | 鞍钢集团矿业有限公司 | Processing method of open-pit mine slope monitoring plane data based on measurement robot |
| CN107369162A (en) * | 2017-07-21 | 2017-11-21 | 华北电力大学（保定） | A kind of generation method and system of insulator candidate target region |
Non-Patent Citations (2)
| Title |
|---|
| ***等: "基于Harris角点的彩色图像文字检测", 《微电子学与计算机》 * |
| 赵振刚: "图像角点检测算法的研究", 《中国优秀硕士学位论文全文数据库 (信息科技辑)》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109829910A (en) * | 2019-02-01 | 2019-05-31 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of PCB defect inspection method searched based on neighborhood |
| CN109829910B (en) * | 2019-02-01 | 2020-10-30 | 佛山市南海区广工大数控装备协同创新研究院 | PCB defect detection method based on neighborhood search |
| CN113310987A (en) * | 2020-02-26 | 2021-08-27 | 保定市天河电子技术有限公司 | Tunnel lining surface detection system and method |
| CN111429450A (en) * | 2020-04-10 | 2020-07-17 | 展讯通信（上海）有限公司 | Corner point detection method, system, equipment and storage medium |
| CN111429450B (en) * | 2020-04-10 | 2022-08-16 | 展讯通信（上海）有限公司 | Corner point detection method, system, equipment and storage medium |
| CN112387982A (en) * | 2020-10-21 | 2021-02-23 | 上海交通大学 | Laser additive process power combined regulation and control method |
| CN112387982B (en) * | 2020-10-21 | 2021-10-12 | 上海交通大学 | Laser additive process power combined regulation and control method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107808378B (en) | Method for detecting potential defects of complex-structure casting based on vertical longitudinal and transverse line profile features | |
| CN111145161B (en) | Pavement crack digital image processing and identifying method | |
| CN109900711A (en) | Workpiece, defect detection method based on machine vision | |
| CN107705288B (en) | Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion | |
| CN113592861B (en) | Bridge crack detection method based on dynamic threshold | |
| CN103077384B (en) | A kind of method and system of vehicle-logo location identification | |
| CN102175700B (en) | Method for detecting welding seam segmentation and defects of digital X-ray images | |
| CN108492306A (en) | A kind of X-type Angular Point Extracting Method based on image outline | |
| CN108009591A (en) | A kind of contact network key component identification method based on deep learning | |
| CN109784344A (en) | A kind of non-targeted filtering method of image for ground level mark identification | |
| CN109993099A (en) | A kind of lane line drawing recognition methods based on machine vision | |
| CN108038883A (en) | A kind of Crack Detection and recognition methods applied to highway pavement video image | |
| CN103345755A (en) | Chessboard angular point sub-pixel extraction method based on Harris operator | |
| CN109376740A (en) | A kind of water gauge reading detection method based on video | |
| CN109685788B (en) | Automatic detection method for floor defect image based on morphological characteristics | |
| CN106887004A (en) | A kind of method for detecting lane lines based on Block- matching | |
| CN109540925B (en) | Complex ceramic tile surface defect detection method based on difference method and local variance measurement operator | |
| CN106815583A (en) | A kind of vehicle at night license plate locating method being combined based on MSER and SWT | |
| CN111932490B (en) | Visual system grabbing information extraction method for industrial robot | |
| CN111539436B (en) | Rail fastener positioning method based on straight template matching | |
| CN105741281B (en) | Method for detecting image edge based on neighborhood dispersion | |
| CN106327464A (en) | Edge detection method | |
| CN104966302B (en) | A kind of detection localization method of any angle laser cross | |
| CN105787912A (en) | Classification-based step type edge sub pixel localization method | |
| CN106709952A (en) | Automatic calibration method of display screen |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180904 |