CN103218822B - Based on the image characteristic point automatic testing method of disappearance importance - Google Patents
Based on the image characteristic point automatic testing method of disappearance importance Download PDFInfo
- Publication number
- CN103218822B CN103218822B CN201310162920.9A CN201310162920A CN103218822B CN 103218822 B CN103218822 B CN 103218822B CN 201310162920 A CN201310162920 A CN 201310162920A CN 103218822 B CN103218822 B CN 103218822B
- Authority
- CN
- China
- Prior art keywords
- importance
- disappearance
- point
- standard deviation
- image
- 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.)
- Expired - Fee Related
Links
Landscapes
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of image characteristic point automatic testing method based on disappearance importance, comprising: gather image, input computing machine and be translated into gray level image; The average disappearance importance at each point place in computed image; The standard deviation disappearance importance at each point place in computed image; Utilize average to lack importance and carry out marginal point mark; Utilize standard deviation to lack importance and carry out unique point mark; Export the image characteristic point marked.Compared to existing method, method provided by the invention has clear superiority in positioning precision.
Description
Technical field
The present invention relates to the automatic detection field of characteristics of image in computer vision, particularly relate to a kind of automatic testing method of image characteristic point.
Background technology
Unique point Automatic Measurement Technique has important application at numerous areas such as image retrieval, object identification, video tracking and augmented realities.There is comparatively multi-characteristic points automatic detection algorithm in the last few years, more representational algorithm comprises: the SUSAN method that (1) paper " SUSAN – ANewApproachtoLowLevelImageProcessing.InternationalJourn alofComputerVision.1997,23 (1): 45-78 " proposes; (2) the CSS method that proposes of paper " RobustImageCornerDetectionThroughCurvatureScaleSpace.IEE ETransonPatternAnalysisandMachineIntelligence.1998,20 (12): 1376-1381 "; (3) paper " based on Corner Detection and the sub-pixel positioning of local direction distribution. Journal of Software .2008,19 (11): 2932-2942 " the LOD method that proposes.
In said method, SUSAN method to image noise and Threshold selection comparatively responsive; Although the unique point that LOD method detects has higher precision, because its step is various and need to carry out data fitting, efficiency is lower; CCS algorithm, due to its excellent combination property, is detection method the most conventional at present.The rudimentary algorithm step of CSS method is: step one, utilizes Canny edge detection operator to carry out rim detection; Step 2, on outline map by interruption incomplete edge conjunction be complete edge; Step 3, outline map upon connection detects curvature maximum point; Step 4, by carrying out at metric space following the tracks of the exact position finding unique point.
The subject matter of the method is the problem that step one uses Canny operator and brings: the gaussian filtering that (1) Canny edge detection operator uses causes picture edge characteristic position to offset, so need step 4 in the exact position of multiscale space tracking characteristics point, realize relative complex on the one hand, the position obtained by tracking is not on the other hand still very accurate; (2) gaussian filtering carried out in Canny operator implementation causes the edge obtained often to rupture and imperfect; so need step 2 edge to re-start connection; connection procedure often can cause characteristic point position to offset, disappearance, mistake, and the accuracy that final effect characteristics detects.
Summary of the invention
The present invention mainly solves the automatic test problems of unique point in digital picture, and object is to provide and does not a kind ofly need to carry out the simple of gaussian filtering and have the unique point automatic testing method of more high accuracy.For realizing this object, method provided by the invention mainly comprises the following steps:
Step S1: gather image, input computing machine and be translated into gray level image;
Step S2: the average disappearance importance at each point place in computed image;
Step S3: the standard deviation disappearance importance at each point place in computed image;
Step S4: utilize average to lack importance and carry out marginal point mark;
Step S5: utilize standard deviation to lack importance and carry out unique point mark;
Step S6: export the unique point that step S5 marks.
Image characteristic point automatic testing method based on disappearance importance provided by the invention, the basis that the basic ideas inheriting CSS algorithm " are first carried out rim detection and then carried out feature point detection on outline map " is improved.Compared to CSS algorithm, the average importance that the method is defined by Corpus--based Method amount and standard deviation lack importance, do not re-use Canny edge detection operator and carry out rim detection, avoid because gaussian filtering causes final edge detection results inaccurate and imperfect, and finally ensure that the accuracy and integrality of on outline map, carrying out feature point detection.In addition, owing to no longer needing to follow the tracks of in the position of metric space to unique point, relative to existing CSS method, method provided by the invention is more simple and be easy to realize.
Accompanying drawing explanation
Figure 1 shows that the process flow diagram of the image characteristic point automatic testing method that the present invention is based on disappearance importance.
Embodiment
Be illustrated in figure 1 the process flow diagram of the image characteristic point automatic testing method that the present invention is based on disappearance importance.The key step of method provided by the invention comprises: gather image, input computing machine and be translated into gray level image; The average disappearance importance at each point place in computed image; The standard deviation disappearance importance at each point place in computed image; Utilize average to lack importance and carry out marginal point mark; Utilize standard deviation to lack importance and carry out unique point mark; Export the image characteristic point marked.The concrete implementation detail of each step is as follows:
Step S1: gather image, input computing machine and be translated into gray level image;
Step S2: the average disappearance importance at each point place in computed image, concrete mode is: for any position X (x in image, y), first will with an X (x, y) centered by, radius is that the border circular areas of R is defined as X (x, y) supporting zone is also designated as Ω (X), then calculates the mean value of each pixel gray-scale value in Ω (X) and is designated as m
1(X), then calculate in Ω (X) remove an X (x, y) afterwards each pixel gray-scale value mean value and be designated as m
2(X), finally by m (X)=| m
1(X)-m
2(X) | be defined as the average disappearance importance at X (x, a y) place, during definition supporting zone, the span of R is 1 ~ 3;
Step S3: the standard deviation disappearance importance at each point place in computed image, concrete mode is: for any position X (x in image, y), first an X (x is determined according to mode described in step S2, y) supporting zone Ω (X), then calculates the standard deviation of each pixel gray-scale value in Ω (X) and is designated as s
1(X), then calculate in Ω (X) remove an X (x, y) afterwards each pixel gray-scale value standard deviation and be designated as s
2(X), finally by s (X)=| s
1(X)-s
2(X) | be defined as the standard deviation disappearance importance at X (x, a y) place, during definition supporting zone, the span of R is 1 ~ 3;
Step S4: utilize average to lack importance and carry out marginal point mark, concrete mode is: first calculated threshold T=kMean (M), wherein Mean (M) represents the average of each point place average disappearance importance in the whole image that step S2 calculates, the span of k is 2 ~ 5, then, if the average disappearance importance at certain some place is greater than T in image, then this point is labeled as marginal point;
Step S5: utilize standard deviation to lack importance and carry out unique point mark, concrete mode is: for the standard deviation disappearance importance at step S3 gained each point place, standard deviation disappearance importance step S4 not being labeled as the position of marginal point corresponding is set to 0, then, if the standard deviation disappearance importance at certain some place is maximal value in 5 × 5 neighborhoods of this point in image, then this point is labeled as unique point;
Step S6: export the unique point that step S5 marks.
Image characteristic point automatic testing method based on disappearance importance provided by the invention, the basis that the basic ideas inheriting CSS algorithm " are first carried out rim detection and then carried out feature point detection on outline map " is improved.Compared to CSS algorithm, the method lacks importance by the average of definition Corpus--based Method amount and standard deviation lacks importance, do not re-use Canny edge detection operator and carry out rim detection, avoid the edge detection results caused due to gaussian filtering inaccurate and imperfect, and finally ensure that the accuracy and integrality of on outline map, carrying out feature point detection.In addition, owing to no longer needing to follow the tracks of in the position of metric space to unique point, relative to existing CSS method, method provided by the invention is more simple and be easy to realize.
Claims (1)
1., based on an image characteristic point automatic testing method for disappearance importance, it is characterized in that, comprise step:
Step S1: gather image, input computing machine and be converted into gray level image;
Step S2: the average disappearance importance calculating each point place in gray level image, concrete mode is: for any position X (x in gray level image, y), first will with an X (x, y) centered by, radius is that the border circular areas of R is defined as X (x, y) supporting zone is also designated as Ω (X), then calculates the mean value of each pixel gray-scale value in Ω (X) and is designated as m
1(X), then calculate in Ω (X) remove an X (x, y) afterwards each pixel gray-scale value mean value and be designated as m
2(X), finally by m (X)=| m
1(X)-m
2(X) | be defined as the average disappearance importance at X (x, a y) place, during definition supporting zone, the span of R is 1 ~ 3;
Step S3: the standard deviation disappearance importance calculating each point place in gray level image, concrete mode is: for any position X (x in gray level image, y), first an X (x is determined according to mode described in step S2, y) supporting zone Ω (X), then calculates the standard deviation of each pixel gray-scale value in Ω (X) and is designated as s
1(X), then calculate in Ω (X) remove an X (x, y) afterwards each pixel gray-scale value standard deviation and be designated as s
2(X), finally by s (X)=| s
1(X)-s
2(X) | be defined as the standard deviation disappearance importance at X (x, a y) place, during definition supporting zone, the span of R is 1 ~ 3;
Step S4: utilize average to lack importance and carry out marginal point mark, concrete mode is: first calculated threshold T=kMean (M), wherein Mean (M) represents the average of each point place average disappearance importance in the whole gray level image that step S2 calculates, and the span of k is 2 ~ 5; Then, if the average disappearance importance at certain some place is greater than T in gray level image, then this point is labeled as marginal point;
Step S5: utilize standard deviation to lack importance and carry out unique point mark, concrete mode is: for the standard deviation disappearance importance at step S3 gained each point place, standard deviation disappearance importance step S4 not being labeled as the position of marginal point corresponding is set to 0, then, if the standard deviation disappearance importance at certain some place is maximal value in 5 × 5 neighborhoods of this point in gray level image, then this point is labeled as unique point;
Step S6: export the unique point that step S5 marks.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310162920.9A CN103218822B (en) | 2013-05-06 | 2013-05-06 | Based on the image characteristic point automatic testing method of disappearance importance |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310162920.9A CN103218822B (en) | 2013-05-06 | 2013-05-06 | Based on the image characteristic point automatic testing method of disappearance importance |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103218822A CN103218822A (en) | 2013-07-24 |
CN103218822B true CN103218822B (en) | 2016-02-17 |
Family
ID=48816565
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310162920.9A Expired - Fee Related CN103218822B (en) | 2013-05-06 | 2013-05-06 | Based on the image characteristic point automatic testing method of disappearance importance |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103218822B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105243661A (en) * | 2015-09-21 | 2016-01-13 | 成都融创智谷科技有限公司 | Corner detection method based on SUSAN operator |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7054506B2 (en) * | 2001-05-29 | 2006-05-30 | Sii Nanotechnology Inc. | Pattern measuring method and measuring system using display microscope image |
CN101980250A (en) * | 2010-10-15 | 2011-02-23 | 北京航空航天大学 | Method for identifying target based on dimension reduction local feature descriptor and hidden conditional random field |
-
2013
- 2013-05-06 CN CN201310162920.9A patent/CN103218822B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7054506B2 (en) * | 2001-05-29 | 2006-05-30 | Sii Nanotechnology Inc. | Pattern measuring method and measuring system using display microscope image |
CN101980250A (en) * | 2010-10-15 | 2011-02-23 | 北京航空航天大学 | Method for identifying target based on dimension reduction local feature descriptor and hidden conditional random field |
Non-Patent Citations (4)
Title |
---|
Harris相关与特征匹配;王旭光等;《模式识别与人工智能》;20090831;第22卷(第4期);505-513 * |
HLD:A robust descriptor for line matching;zhihengwang等;《11th IEEE International Conference on Computer-Aided Design and Computer Graphics》;20090821;128-133 * |
均值-标准差描述子与直线匹配;王志衡等;《模式识别与人工智能》;20090215;第22卷(第1期);32-39 * |
基于亮度序的均值标准差描述子;王志衡等;《模式识别与人工智能》;20130415;第26卷(第4期);409-416 * |
Also Published As
Publication number | Publication date |
---|---|
CN103218822A (en) | 2013-07-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5699788B2 (en) | Screen area detection method and system | |
CN102831582B (en) | A kind of depth image of Microsoft somatosensory device Enhancement Method | |
JP6435750B2 (en) | Three-dimensional coordinate calculation apparatus, three-dimensional coordinate calculation method, and three-dimensional coordinate calculation program | |
JP6842039B2 (en) | Camera position and orientation estimator, method and program | |
CN103727930B (en) | A kind of laser range finder based on edge matching and camera relative pose scaling method | |
CN107341802A (en) | It is a kind of based on curvature and the compound angular-point sub-pixel localization method of gray scale | |
CN102982545B (en) | A kind of image depth estimation method | |
CN111754536B (en) | Image labeling method, device, electronic equipment and storage medium | |
CN102589435A (en) | Efficient and accurate detection method of laser beam center under noise environment | |
US20150090793A1 (en) | Method and system for determining edge line in qr code binary image | |
CN103400141A (en) | Method for calculating thickness of ice coated on transmission line on basis of improved image method | |
US11488354B2 (en) | Information processing apparatus and information processing method | |
CN105069453A (en) | Image correction method and apparatus | |
CN103886597A (en) | Circle detection method based on edge detection and fitted curve clustering | |
CN109389165A (en) | Oil level gauge for transformer recognition methods based on crusing robot | |
CN104036516A (en) | Camera calibration checkerboard image corner detection method based on symmetry analysis | |
CN108573251A (en) | Character area localization method and device | |
CN102881027A (en) | Method and system for detecting quadrangle of given region in image | |
CN104077775A (en) | Shape matching method and device combining skeleton feature points and shape context | |
WO2020093566A1 (en) | Cerebral hemorrhage image processing method and device, computer device and storage medium | |
CN105335960A (en) | Image segmentation method combining edge detection algorithm with watershed algorithm | |
CN108876771A (en) | A kind of detection method of undercut welding defect | |
CN103837135B (en) | Workpiece inspection method and system thereof | |
CN113762397B (en) | Method, equipment, medium and product for training detection model and updating high-precision map | |
CN103218822B (en) | Based on the image characteristic point automatic testing method of disappearance importance |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160217 |