CN101877063A - Sub-pixel characteristic point detection-based image matching method - Google Patents
Sub-pixel characteristic point detection-based image matching method Download PDFInfo
- Publication number
- CN101877063A CN101877063A CN2009102415427A CN200910241542A CN101877063A CN 101877063 A CN101877063 A CN 101877063A CN 2009102415427 A CN2009102415427 A CN 2009102415427A CN 200910241542 A CN200910241542 A CN 200910241542A CN 101877063 A CN101877063 A CN 101877063A
- Authority
- CN
- China
- Prior art keywords
- image
- point
- detected
- sampled
- gaussian
- 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
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000001514 detection method Methods 0.000 title claims abstract description 18
- 238000012887 quadratic function Methods 0.000 claims abstract description 14
- 239000011159 matrix material Substances 0.000 claims abstract description 10
- 238000001914 filtration Methods 0.000 claims abstract description 6
- 230000015572 biosynthetic process Effects 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 3
- 238000012986 modification Methods 0.000 abstract 1
- 230000004048 modification Effects 0.000 abstract 1
- 230000000750 progressive effect Effects 0.000 abstract 1
- 238000005070 sampling Methods 0.000 abstract 1
- 230000006870 function Effects 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 6
- 238000006243 chemical reaction Methods 0.000 description 5
- 230000005764 inhibitory process Effects 0.000 description 4
- 230000008447 perception Effects 0.000 description 4
- 238000005286 illumination Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000000605 extraction Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000003190 augmentative effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a sub-pixel characteristic point detection-based image matching method. The method comprises the following steps of: performing down-sampling on an image to be detected until the smaller one of the length and the width of the image to be detected is less than 8 pixels, acquiring and performing geometric progression variance Gaussian filtering on each down-sampled image to obtain continuous and progressive Gaussian blur images of which the maximum value of the variance is 2, establishing and acquiring the Gaussian pyramid of the image to be detected, evaluating a second derivative Ixx in x direction, a second derivative Iyy in y direction and a second derivative Ixy in xy direction; calculating the Hessian matrix of the Gaussian pyramid of the image to be detected and all local maximum value points of a Harris function value; and fitting a three-dimensional quadratic function with each local maximum value point on the Harris function value and 26 neighborhood points around each local maximum value point, updating the position of an extreme point of the quadratic function by using the floating number position of the extreme point, performing sub-pixel accuracy positioning modification on the position of the extreme point and outputting all characteristic points.
Description
Technical field
The invention belongs to area of pattern recognition, relate to technology such as Flame Image Process and computer vision, particularly relate to feature detection, object identification, target following etc.
Background technology
The characteristics of image detection is the basic problem in the computer vision, and the characteristics of image detection algorithm of stability and high efficiency provides solid underlying basis for the solution of other problems.Along with the development of technology and the reduction gradually of hardware device price, camera and video camera have become one of equipment of widespread use in people's daily life.People are converted into three-dimensional perception to the requirement of scene perception from original two-dimentional perception, i.e. the three-dimensional configuration of object and the 3 d pose of scene in the perception real world.Because the variation of object is various in the real world, actual object is directly carried out three-dimensional modeling or scene is carried out the space demarcation having consumed the lot of manpower and material resources resource.To the three-dimensional body modeling or three-dimensional scenic is rebuild is one of forward position research direction that receives much concern in recent years, its detects from the image that camera or video camera are caught, identification, follow the tracks of object and scene and the attitude in its behavior three dimensions is estimated by image.Although present existing computerized three-dimensional reconstruction technology is used widely, wherein several important problem still are worth inquiring into.These problems are accuracys, stability and high efficiency.Therefore, as the bottom problem in the computer vision, the research that image characteristic point detects on these three problems is particularly important.At these three problems, develop a cover accuracy height, good stability can carry out the characteristics of image detection algorithm in real time, and place mat is carried out in the application of reality.
There is point in the image with good location ability and separating capacity.People can find out these representative points easily from image.But concerning computing machine, the detection of unique point is a very difficult problem in the image.
These difficult problems generally can be summed up as the following aspects: the conversion of illumination, and the table at visual angle changes, and it is fuzzy that video camera causes, variation that the object attitude causes or the like.In recent years, the concern of extensively being sent out by people at the method for visual angle change.The conversion of graphical rule has very important researching value as one of basic conversion of image in computer vision.The feature detection that change of scale carried out at image has very important meaning to this problem.
Summary of the invention
The constant feature detection subbase of existing graphical rule is in the local Gaussian model of image; For asking for of the constant unique point of yardstick stable in the image, generally be to be based upon on the second order difference matrix of image gaussian pyramid; But unsettled characteristics of image such as edge has bigger negative effect to the extraction of image characteristic point, how effectively to avoid unsettled unique points such as edge that very important Research Significance is arranged in image detection; In order to solve prior art problems, the purpose of this invention is to provide a kind of image matching method that detects based on the constant sub-pixel characteristic point of the yardstick of image local Gauss model.
To achieve these goals, the invention provides a kind of image matching method based on the sub-pixel level image feature point detection, this method step comprises:
Step S1: treat detected image and carry out down-sampled, up to the length of image to be detected and wide in less one less than 8 pixels, obtain down-sampled image, again each down-sampled image is carried out the gaussian filtering of Geometric Sequence variance, obtain the image of the continuous Gaussian Blur that goes forward one by one, the maximal value of variance is 2, sets up and obtain the gaussian pyramid of image to be detected;
Step S2: the gaussian pyramid of setting up image to be detected is asked for the second derivative Ixx of x direction, ask for the second derivative Iyy of y direction, ask for the second derivative Ixy of xy direction;
Step S3: Hai Sen (Hessian) matrix of asking for the gaussian pyramid of image to be detected is: [Ixx, Ixy; Ixy, Iyy], ask for Harris (Harris) functional value and be C=Ixx * Iyy-Ixy * Ixy-k * (Ixx+Iyy)
2, wherein the k span is [0.04-0.16];
Step S4: all local maximum points of asking for the gaussian pyramid Harris functional value of image to be detected;
Step S5: (3 * 3 * 3-1) neighborhoods are put the quadratic function of common match three-dimensional to treat each local maximum point on the Harris functional value of gaussian pyramid of detected image and 26 around it, and with the position of this extreme point of floating number position renewal of the extreme point of quadratic function, the correction of sub-pixel precision location is carried out in the extreme point position, export all unique points at last.
Wherein, the concrete steps of the down-sampled image of described acquisition are as follows:
Step S11: treat detected image and carry out smothing filtering and carry out noise remove, obtain the image of denoising;
Step S12: the image of denoising is carried out down-sampled, obtain down-sampled image;
Step S13: to down-sampled image repeating step S11 and step S12, minimum elongated up to down-sampled image less than 8 pixels.
Wherein, ask for local maximum point step:
Step S31: whether the value that detects each point in the gaussian pyramid of image to be detected is the maximum value in its 26=3 * 3 * 3-1 neighborhood, if this point coordinate is the initial coordinate of a unique point;
Step S32: the unique point in the Gaussian image pyramid that all are to be detected deposits the unique point formation in.
Wherein, the concrete steps of sub-pixel precision location are as follows:
Step S41: each the unique point x in the unique point formation and its be the point of 26 neighborhoods on every side, and quadratic function of match is in three dimensions:
D is an abstract three-dimensional function, and x is the variable of this three-dimensional function, and wherein T is transpose of a matrix;
Step S42: the position of asking the extreme point of this quadratic function itself
And upgrade the coordinate of original unique point x, wherein
The extreme point of three-dimensional function.
Beneficial effect of the present invention: method of the present invention is that the variation at graphical rule can access abundant point-of-interest, so as after processing in obtain a large amount of image informations.The point that the inventive method is sought in metric space has extremely strong stability, and to noise, influences such as illumination have antijamming capability.It is bottom problem in the computer vision problem that characteristics of image detects, and generally as the bottom-up information support of additive method, supports the upper strata algorithm, and to have separating capacity strong for the upper strata algorithm provides, and has abundant semantic unique point.Method of the present invention is to be easy to realize and use, and mainly can be applied to following several aspect:
(1) based on the supervisory system of signature tracking, help system is followed the tracks of the target in the scene.Obtain the behavior semanteme of point-of-interest in the scene, thereby the incident in the scene is understood.
(2) based on the three-dimensional body modeling of Image Feature Point Matching, be used for the three-dimensional body of complexity is carried out morphological analysis and three-dimensional reconstruction, virtual reality there is great function
(3) based on the augmented reality system of signature tracking, be used for by the tracking of image static nature point being obtained the three-dimensional spatial information of scene, thus the content by virtual object enhanced scene.
(4) based on the object identification and the classification of local feature, the more unsettled scene of global information is carried out local feature extract, can effectively tackle and block change in location, problems such as dimensional variation.
Description of drawings
Fig. 1 illustrates the structure result based on image pyramid;
Fig. 2 illustrates Harris functional value result;
Fig. 3 illustrates 26 neighborhoods and asks local maximum;
Fig. 4 is the process flow diagram of the inventive method.
Embodiment
Describe each related detailed problem in the technology of the present invention method in detail below in conjunction with accompanying drawing.Feature detection algorithm based on the image local Gauss model has improved the stability that characteristics of image detects, the rich and accuracy of feature.Utilize of the inhibition of Harris Harris function to the edge, the present invention has realized a sub-pixel level image feature detection and matching system, the foundation of image pyramid is shown as Fig. 1, trying to achieve of the image gloomy Harris functional value in sea shown in Figure 2, Fig. 3 illustrates 26 neighborhoods and asks local maximum, method of the present invention is set up the yardstick pyramid of image based on image; Based on the Hessian matrix, ask for the Harris functional value; Carry out non-maximum the inhibition, ask for stable extreme point; Carry out sub-pixel characteristic point and accurately locate, as Fig. 4 the flow process of the inventive method is shown, concrete grammar is as follows:
Set up the gaussian pyramid step of image: treat detected image and carry out down-sampled, up to the length of image and wide in less one less than 8 pixels, each down-sampled image is carried out the gaussian filtering of Geometric Sequence variance, obtain the image of the continuous Gaussian Blur that goes forward one by one, the maximal value of variance is 2, sets up and obtain the gaussian pyramid of image to be detected.
Ask for the Hessian matrix and ask Harris functional value pyramid step: the gaussian pyramid of the image to be detected set up is asked for the second derivative Ixx of x direction, ask for the second derivative Iyy of y direction, ask for the second derivative Ixy of xy direction.The Hessian matrix of then asking for the gaussian pyramid of image to be detected is: [Ixx, Ixy; Ixy, Iyy].Ask for the Harris functional value and be C=Ixx * Iyy-Ixy * Ixy-k * (Ixx+Iyy)
2, wherein the k span is [0.04-0.16].
Ask the local maximum step: Harris functional value pyramid is asked all local maximum points, local maximum point is added the unique point formation; Described 26 neighborhoods are the neighborhoods in 3 dimension spaces.
Sub-pixel precision positioning step S4: to each local maximum point of the Harris functional value of the gaussian pyramid for the treatment of detected image or claim unique point and the quadratic function of common match 3 dimension spaces of 26 points around it, and, the extreme point position is carried out the correction of sub-pixel precision location with position, the floating number level position of the extreme point of quadratic function as extreme point itself.26 points around described each unique point are the neighbor points in 3 dimension spaces.
The hardware minimalist configuration that method of the present invention needs is: P43.0G CPU, the computing machine of 512M internal memory.On the hardware of this configuration level, adopt the C Plus Plus programming to realize this method.The committed step that method of the present invention is related to describes in detail one by one below, and the basic step in the method for the present invention is identical, and concrete form is as described below:
At first, the foundation of image gaussian pyramid:
Original image size is 316 * 198 pixels.The extraction of characteristics of image mainly is to obtain by obtaining to have the stable point of gradient.Stable point is the target that we are concerned about in characteristic dimension space and the image coordinate space.In order to obtain the information in graphical rule space, we need set up the gaussian pyramid of image, carry out progressively Gaussian Blur of image by the mode of iteration, set up image pyramid, the variance of Gaussian Blur from 0.5 to 2.And image is carried out down-sampled, the image after down-sampled is carried out progressively Gaussian Blur equally, finally obtain the image on all yardsticks.
Image after down-sampled is repeated down-sampled according to the method for original image, obtain the image that one group of size is successively decreased, these images are carried out progressively Gaussian Blur equally, as shown in Figure 1.Down-sampled minimum one deck size guarantees that the minimum of image becomes the length of side and gets final product greater than 8 pixels
Its two, set up Harris fundamental function value pyramid:
Image to all foundation among Fig. 1 is asked for x, y, and the second derivative of three directions of xy, and use the Harris function to try to achieve functional value: C=Ixx * Iyy-Ixy * Ixy-k * (Ixx+Iyy)
2The result as shown in Figure 2.
Its three, ask local maximum to obtain the stable characteristics point:
Each point in all images is asked for non-maximum the inhibition, is that the point of local maximum is saved in the formation and goes with all.Employed 26 neighborhoods of non-maximum inhibition as shown in Figure 3.
Its four, ask for the sub-pixel location of local maximum point:
To each point in the Harris image pyramid and 6 pixels (three-dimensional 26 neighborhoods) match quadratic function of being adjacent
D is an abstract three-dimensional function, and x is the variable of this three-dimensional function, asks the extreme point of this quadratic function:
Extreme point with three-dimensional function
Upgrade the coordinate of original unique point x, make x accurately turn to floating number from integer.This is the sub-pixel precision level position of all unique points, and wherein T is transpose of a matrix.
In a word, the present invention proposes a kind of simple and effective sub-pixel characteristic point detection algorithm based on the image local Gauss model.Carry out a large amount of experiments on some databases in the world, verified the validity and the stability of our methods.The present invention is easy to realize, stable performance.The present invention can improve the stability that image characteristic point detects, and the sub-pixel characteristic point that the present invention detected is to rotation, illumination, and conversion such as yardstick all have very strong robustness, and to image mosaic, the high layer methods of computer vision such as tracking have good guarantee.
The above; only be the embodiment among the present invention; but protection scope of the present invention is not limited thereto; anyly be familiar with the people of this technology in the disclosed technical scope of the present invention; can understand conversion or the replacement expected; all should be encompassed in of the present invention comprising within the scope, therefore, protection scope of the present invention should be as the criterion with the protection domain of claims.
Claims (4)
1. the image matching method of a sub-pixel level image feature point detection is characterized in that:
Step S1: treat detected image and carry out down-sampled, up to the length of image to be detected and wide in less one less than 8 pixels, obtain down-sampled image, again each down-sampled image is carried out the gaussian filtering of Geometric Sequence variance, obtain the image of the continuous Gaussian Blur that goes forward one by one, the maximal value of variance is 2, sets up and obtain the gaussian pyramid of image to be detected;
Step S2: the gaussian pyramid of setting up image to be detected is asked for the second derivative Ixx of x direction, ask for the second derivative Iyy of y direction, ask for the second derivative Ixy of xy direction;
Step S3: Hai Sen (Hessian) matrix of asking for the gaussian pyramid of image to be detected is: [Ixx, Ixy; Ixy, Iyy], ask for Harris (Harris) functional value and be C=Ixx * Iyy-Ixy * Ixy-k * (Ixx+Iyy)
2, wherein the k span is [0.04-0.16];
Step S4: all local maximum points of asking for the gaussian pyramid Harris functional value of image to be detected;
Step S5: (3 * 3 * 3-1) neighborhoods are put the quadratic function of common match three-dimensional to treat each local maximum point on the Harris functional value of gaussian pyramid of detected image and 26 around it, and with the position of this extreme point of floating number position renewal of the extreme point of quadratic function, the correction of sub-pixel precision location is carried out in the extreme point position, export all unique points at last.
2. by the image matching method of the described image characteristic point detection of claim 1, it is characterized in that: the concrete steps of the down-sampled image of described acquisition are as follows:
Step S11: treat detected image and carry out smothing filtering and carry out noise remove, obtain the image of denoising;
Step S12: the image of denoising is carried out down-sampled, obtain down-sampled image;
Step S13: to down-sampled image repeating step S11 and step S12, minimum elongated up to down-sampled image less than 8 pixels.
3. by the image matching method of the described image characteristic point detection of claim 1, it is characterized in that: ask for local maximum point step:
Step S31: whether the value that detects each point in the gaussian pyramid of image to be detected is the maximum value in its 26=3 * 3 * 3-1 neighborhood, if this point coordinate is the initial coordinate of a unique point;
Step S32: deposit the unique point in the gaussian pyramid of all images to be detected in the unique point formation.
4. by the image matching method of the described image characteristic point detection of claim 1, it is characterized in that: the concrete steps of sub-pixel precision location are as follows:
Step S41: each the unique point x in the unique point formation and its be the point of 26 neighborhoods on every side, and quadratic function of match is in three dimensions:
D is an abstract three-dimensional function, and x is the variable of this three-dimensional function, and wherein T is transpose of a matrix;
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009102415427A CN101877063A (en) | 2009-11-25 | 2009-11-25 | Sub-pixel characteristic point detection-based image matching method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009102415427A CN101877063A (en) | 2009-11-25 | 2009-11-25 | Sub-pixel characteristic point detection-based image matching method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101877063A true CN101877063A (en) | 2010-11-03 |
Family
ID=43019615
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2009102415427A Pending CN101877063A (en) | 2009-11-25 | 2009-11-25 | Sub-pixel characteristic point detection-based image matching method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101877063A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2014040281A1 (en) * | 2012-09-14 | 2014-03-20 | 华为技术有限公司 | Augmented reality processing method and device for mobile terminal |
CN104168820A (en) * | 2012-03-14 | 2014-11-26 | 富士胶片株式会社 | Image processing device, method, and program |
CN104515502A (en) * | 2013-09-28 | 2015-04-15 | 沈阳新松机器人自动化股份有限公司 | Robot hand-eye stereo vision measurement method |
CN108027248A (en) * | 2015-09-04 | 2018-05-11 | 克朗设备公司 | The industrial vehicle of positioning and navigation with feature based |
CN108573269A (en) * | 2017-10-24 | 2018-09-25 | 北京金山云网络技术有限公司 | Image characteristic point matching method, coalignment, electronic equipment and storage medium |
CN110326027A (en) * | 2017-01-24 | 2019-10-11 | 深圳市大疆创新科技有限公司 | The method and system of signature tracking is carried out using image pyramid |
CN110335292A (en) * | 2019-07-09 | 2019-10-15 | 北京猫眼视觉科技有限公司 | It is a kind of to track the method and system for realizing simulated scenario tracking based on picture |
CN112365439A (en) * | 2020-09-24 | 2021-02-12 | 南昌大学 | Method for synchronously detecting forming characteristics of GMAW (gas metal arc welding) weld of galvanized steel and welding gun direction in real time |
CN113223066A (en) * | 2021-04-13 | 2021-08-06 | 浙江大学 | Multi-source remote sensing image matching method and device based on characteristic point fine tuning |
-
2009
- 2009-11-25 CN CN2009102415427A patent/CN101877063A/en active Pending
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104168820A (en) * | 2012-03-14 | 2014-11-26 | 富士胶片株式会社 | Image processing device, method, and program |
CN104168820B (en) * | 2012-03-14 | 2016-05-11 | 富士胶片株式会社 | Image processing apparatus and method |
WO2014040281A1 (en) * | 2012-09-14 | 2014-03-20 | 华为技术有限公司 | Augmented reality processing method and device for mobile terminal |
CN103814382A (en) * | 2012-09-14 | 2014-05-21 | 华为技术有限公司 | Augmented reality processing method and device of mobile terminal |
CN103814382B (en) * | 2012-09-14 | 2016-10-05 | 华为技术有限公司 | The augmented reality processing method and processing device of mobile terminal |
CN104515502A (en) * | 2013-09-28 | 2015-04-15 | 沈阳新松机器人自动化股份有限公司 | Robot hand-eye stereo vision measurement method |
CN108027248A (en) * | 2015-09-04 | 2018-05-11 | 克朗设备公司 | The industrial vehicle of positioning and navigation with feature based |
CN110326027A (en) * | 2017-01-24 | 2019-10-11 | 深圳市大疆创新科技有限公司 | The method and system of signature tracking is carried out using image pyramid |
CN108573269A (en) * | 2017-10-24 | 2018-09-25 | 北京金山云网络技术有限公司 | Image characteristic point matching method, coalignment, electronic equipment and storage medium |
CN108573269B (en) * | 2017-10-24 | 2021-02-05 | 北京金山云网络技术有限公司 | Image feature point matching method, matching device, electronic device and storage medium |
CN110335292A (en) * | 2019-07-09 | 2019-10-15 | 北京猫眼视觉科技有限公司 | It is a kind of to track the method and system for realizing simulated scenario tracking based on picture |
CN110335292B (en) * | 2019-07-09 | 2021-04-30 | 北京猫眼视觉科技有限公司 | Method, system and terminal for realizing simulation scene tracking based on picture tracking |
CN112365439A (en) * | 2020-09-24 | 2021-02-12 | 南昌大学 | Method for synchronously detecting forming characteristics of GMAW (gas metal arc welding) weld of galvanized steel and welding gun direction in real time |
CN112365439B (en) * | 2020-09-24 | 2023-10-03 | 南昌大学 | Method for synchronously detecting forming characteristics of GMAW welding seam of galvanized steel and direction of welding gun in real time |
CN113223066A (en) * | 2021-04-13 | 2021-08-06 | 浙江大学 | Multi-source remote sensing image matching method and device based on characteristic point fine tuning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101877063A (en) | Sub-pixel characteristic point detection-based image matching method | |
CN104574347B (en) | Satellite in orbit image geometry positioning accuracy evaluation method based on multi- source Remote Sensing Data data | |
CN101777129B (en) | Image matching method based on feature detection | |
CN112254656B (en) | Stereoscopic vision three-dimensional displacement measurement method based on structural surface point characteristics | |
CN105551028A (en) | Method and system for dynamically updating geographic space data based on remote sensing image | |
CN105550691B (en) | Adaptive mountain valley ridge line extracting method and system based on scale space | |
CN105608667A (en) | Method and device for panoramic stitching | |
CN103279939A (en) | Image stitching processing system | |
CN107341470B (en) | Power transmission line detection method based on aerial images | |
CN109741446B (en) | Method for dynamically generating fine coast terrain by three-dimensional digital earth | |
CN108257155B (en) | Extended target stable tracking point extraction method based on local and global coupling | |
CN102034115B (en) | Markov random field model and non-local prior based image registration method | |
CN107180436A (en) | A kind of improved KAZE image matching algorithms | |
CN102169581A (en) | Feature vector-based fast and high-precision robustness matching method | |
Uchiyama et al. | Toward augmenting everything: Detecting and tracking geometrical features on planar objects | |
CN112634335A (en) | Method for extracting characteristic point pairs of robust remote sensing image facing to nonlinear radiation distortion | |
Huang et al. | Correlation and local feature based cloud motion estimation | |
CN111461046B (en) | Automatic coast erosion and siltation identification method and device based on shoreline data | |
CN105608689B (en) | A kind of panoramic mosaic elimination characteristics of image error hiding method and device | |
CN114596500A (en) | Remote sensing image semantic segmentation method based on channel-space attention and DeeplabV3plus | |
CN115861247A (en) | High-resolution remote sensing image contour multistage regularization method, system and application | |
CN101196988B (en) | Palm locating and center area extraction method of three-dimensional palm print identity identification system | |
CN105678720A (en) | Image matching judging method and image matching judging device for panoramic stitching | |
Miao et al. | A novel marker for robust and accurate phase-based 2D motion estimation from noisy image data | |
CN104933719A (en) | Method for detecting image edge by integral image interblock distance |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20101103 |