CN104167003A - Method for fast registering remote-sensing image - Google Patents
Method for fast registering remote-sensing image Download PDFInfo
- Publication number
- CN104167003A CN104167003A CN201410434279.4A CN201410434279A CN104167003A CN 104167003 A CN104167003 A CN 104167003A CN 201410434279 A CN201410434279 A CN 201410434279A CN 104167003 A CN104167003 A CN 104167003A
- Authority
- CN
- China
- Prior art keywords
- remote sensing
- point
- points
- sensing image
- registered
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012952 Resampling Methods 0.000 claims abstract description 7
- 230000009466 transformation Effects 0.000 claims description 16
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000001514 detection method Methods 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 8
- 238000012163 sequencing technique Methods 0.000 claims description 6
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 3
- 230000005764 inhibitory process Effects 0.000 claims description 2
- 230000009977 dual effect Effects 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 238000000605 extraction Methods 0.000 description 4
- 230000004927 fusion Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to a method for fast registering a remote-sensing image. The method comprises the following steps that S1, ORB feature points are extracted from the remote-sensing image to be registered and a reference remote-sensing image; S2, initial matching is carried out on the extracted ORB feature points, and wrongly-matched feature points are removed from the initially-matched feature points; S3, parameter solving is carried out on the remote-sensing image to be registered; S4, resampling is carried out on the remote-sensing image to be registered, and image registration is completed. The method is beneficial to improving the speed and precision of image registration.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method for quickly registering more than two remote sensing images with different time, different visual angles and different sensors.
Background
Image registration is an important subject in image processing technology, and image registration is widely used in various fields such as remote sensing image analysis, medical image analysis, image fusion, machine vision and other fields at present.
The image registration is a basic problem of remote sensing image processing, and is a premise and a basis for remote sensing image integration analysis and application, such as image data fusion, dynamic change monitoring and the like. In order to timely and accurately monitor the change of the detected area, the obtained remote sensing image needs to be registered in real time. Commonly used image registration methods are generally classified into two categories: a grayscale-based registration method, and a feature-based (e.g., feature point, feature curve) registration method. The registration method based on the gray scale measures the similarity between images by utilizing the gray scale value of the images, the method is simple to implement, but the speed is low, and the remote sensing images are obtained by different sensors, different visual angles, different weathers and the like, so that the algorithm cannot correctly register the images, and the method based on the local invariant features is not easily influenced. The feature-based registration method determines registration parameters according to the geometric relation of important features between the registered images, can reduce the data amount of processing, and has certain robustness for image distortion, noise and the like. The quality of the matching performance depends to a large extent on the method of feature description and the quality of feature extraction.
At present, SIFT extraction feature points are mainly used as registration features for the situations of size scaling, rotation and translation of images. The SIFT features need to establish feature point vectors, and have high dimension and large calculation amount, so that the implementation registration of the remote sensing images cannot be met.
The Fast corner detects a circle of pixel values around the candidate feature point based on the gray value of the image around the feature point, if the gray value difference between enough pixel points in the field around the candidate point and the candidate point is large enough, the candidate point is considered as a feature point, namely:
wherein I (x) is the gray scale of any point on the circumference, I (p) is the gray scale of the center of the circle,ε d for a threshold value of the gray value difference, if N is greater than a given threshold value, typically three quarters of the surrounding circle points, then p is considered a feature point.
The BRIEF descriptor is that in a 31 x 31 pixel block with the size after the image is smoothed, N (N =128, 256, 512) groups of Gaussian distribution random point pixel pairs are selected, and a binary string is formed by comparing the sizes of the pixel pairs and following the criterion of 1 and 0.
Binary detection is defined as:
wherein,p(x 1)、p(x 2) Respectively in pixel block px 1Andx 2pixel gray value of the location.
The BRIEF feature is defined as an n-dimensional binary string vector,namely:
the ORB algorithm has good rotation invariance, characteristic points are extracted by Fast angular points, the Fast angular points are a very quick angular point extraction method, and the Fast angular points do not have rotation invariance, so the ORB adds direction information to the angular points of the ORB by using a centroid method, and the characteristics have rotation invariance. In addition, in order to adapt the features to the case of size scaling, by a method of establishing an image pyramid, feature extraction can be performed in the case of non-uniform image sizes. In addition, when the characteristic point descriptors are established, the Brief descriptors are used for describing, and the Brief descriptors are binary descriptors and can be matched very quickly. Image registration based on ORB features is therefore suitable for real-time registration.
Disclosure of Invention
The invention aims to provide a rapid remote sensing image registration method which is beneficial to improving the speed and the precision of image registration.
In order to achieve the purpose, the technical scheme of the invention is as follows: a method for rapidly registering remote sensing images comprises the following steps:
step S1: respectively extracting ORB characteristic points of the remote sensing image to be registered and the reference remote sensing image;
step S2: performing initial matching on the extracted ORB characteristic points, and eliminating characteristic points which are in error matching from the initially matched characteristic points;
step S3: carrying out parameter solution on the remote sensing image to be registered;
step S4: and resampling the remote sensing image to be registered to finish image registration.
Further, in step S1, an image pyramid is established for the remote sensing image to be registered and the reference remote sensing image, and ORB feature points are extracted for each layer of image pyramid.
Further, in step S1, the method for extracting ORB feature points includes: FAST angular point detection is firstly carried out, Harris angular point detection is carried out, the first N best points are selected, then non-maximum value inhibition is used for verifying angular points, and false edge points are removed, so that the feature points are distributed uniformly.
Further, in step S2, the extracted ORB feature points are initially matched by using a double-threshold-based hamming distance feature point matching method, and feature points that are initially matched are rejected by using an angular point direction included angle as a constraint condition, and the specific method is as follows:
the angular point direction of the characteristic point is obtained by a gray scale centroid method, namely, the angular point direction is represented by the vector direction formed by the angular point and the centroid by calculating the centroid of the gray scale of the pixel in the circular neighborhood of the angular point;
defining the angular point circular neighborhood moment as:the centroid of the angular point circular neighborhood is:and then the vector direction formed by the angular point and the mass center is the angular point direction of the feature point:；
wherein m ispqRepresenting the order moment of p + q, I (x, y) representing the gray value of the pixel point (x, y) in the angular point circular neighborhood, (x, y) representing the coordinate of the pixel point in the angular point circular neighborhood, m00Denotes the zero order moment, m10And m01Each represents an order moment;
Δθ i for the direction of an angular point, Delta, of a feature point in the remote sensing image to be registeredθ i ’Is prepared from radix GinsengWhen the angular point direction of the corresponding characteristic point in the remote sensing image is considered, the included angle of the angular point direction is deltaθ i =Δθ i -Δθ i ’；
Then, the extracted ORB characteristic points are initially matched according to the following steps, and characteristic points which are wrongly matched are eliminated:
a. establishing an rBRIEF descriptor for the corner points extracted by the remote sensing image to be registered and the reference remote sensing image, setting the characteristic point set of the remote sensing image to be registered as { a1, a2, …, an1}, and setting the characteristic point set of the reference remote sensing image as { b1, b2, …, bn2 };
b. respectively comparing the descriptor of the characteristic point a1 of the remote sensing image to be registered with the descriptors of all the characteristic points b1, b2, … and bn2 of the reference remote sensing image, calculating the Hamming distance between the descriptor of a1 and the descriptors of b1, b2, … and bn2, selecting the point with the shortest Hamming distance from the descriptors of b1, b2, … and bn2, calculating the ratio of the shortest Hamming distance to the next shortest Hamming distance, if the ratio is smaller than a set larger threshold value, keeping the point with the shortest Hamming distance and the a1 as initial matching points, and calculating the direction included angle delta of the angular point of the initial matching points and the angular point delta of the initial matching pointsθ 1Otherwise, abandoning;
c. according to the method, the initial matching points of the characteristic points a2, … and an1 of the remote sensing image to be registered in the reference remote sensing image are sequentially solved, and the corresponding angular point direction included angle delta is calculatedθ 2, …, Δθ n ；
d. Sequencing the initial matching points according to the ratio of the shortest Hamming distance to the second shortest Hamming distance from large to small, namely sequencing the matching quality of the feature points from good to poor, and extracting the initial matching points of which the ratio of the shortest Hamming distance to the second shortest Hamming distance is smaller than a set smaller threshold;
e. d, solving the optimal angular point direction included angle delta of the initial matching points extracted in the step d by a least square methodθ m ；
f. Will step withThe angular point direction included angles of all initial paired points obtained in the steps b and c are calculated according to the optimal angular point direction included angle deltaθ m For the constraint, the deviation Δθ m A range of initial matched point culling.
Further, in step S3, the method for performing parameter solution on the remote sensing image to be registered includes:
let the pixel points of the reference remote sensing image bef(x, y) The pixel point of the image to be registered isg(x’, y’) Assuming that the coordinates of a point on the reference remote sensing image are (x i , y i ) The coordinates of the corresponding points on the remote sensing image to be registered are (x i ’, y i ’) Then (1)x i , y i ) And (a)x i ’, y i ’) The affine transformation between is represented as:
in the formula,sis a scale factor, and is a function of,θis a rotation angle, ΔxAnd ΔyRespectively the translation amount of two coordinate axes;
after m characteristic points are obtained, an optimal transformation matrix is solved according to a RANSAC algorithm, namely, registration parameters are determineds、θ、Δx，Δy；
And resampling the remote sensing image to be registered according to the registration parameters, namely finishing image registration.
The method has the advantages that aiming at the conditions of different deformation, illumination and the like existing among remote sensing images, in order to reflect the dynamic change of a monitoring area in time, the method for quickly registering the remote sensing images is provided, mismatching points can be effectively eliminated, the image registration precision is ensured, and the method has strong practicability and wide application prospect.
Drawings
FIG. 1 is a flow chart of an implementation of an embodiment of the present invention.
Fig. 2 is a reference remote sensing image in the embodiment of the present invention.
Fig. 3 is a remote sensing image to be registered in the embodiment of the present invention.
Fig. 4 is an initial feature point matching diagram in the embodiment of the present invention.
Fig. 5 is a feature point matching graph after the mismatching points are removed in the embodiment of the present invention.
FIG. 6 is a diagram of the remote sensing image to be registered after registration and the reference remote sensing image after fusion in the embodiment of the invention.
Detailed Description
The invention is described in further detail below with reference to the figures and the embodiments.
The invention discloses a rapid registration method of remote sensing images, which comprises the following steps as shown in figure 1:
step S1: and respectively establishing an image pyramid for the remote sensing image to be registered (figure 3) and the reference remote sensing image (figure 2), and extracting ORB characteristic points of each layer of image pyramid. The method for extracting ORB characteristic points comprises the following steps: the method comprises the steps of firstly, performing FAST corner detection, wherein the number of corners extracted by FAST is too large, and the corners comprise edge points and pseudo corners, performing Harris corner detection by using a stable corner detector as a Harris corner detection algorithm, selecting the first N best points, then verifying the corners by using non-maximum suppression, and rejecting the pseudo edge points to enable the feature points to be distributed uniformly.
Step S2: in order to obtain the feature points which are matched correctly as much as possible, the extracted ORB feature points are initially matched by adopting a double-threshold-based Hamming distance feature point matching method, the initially matched feature points are subjected to constraint conditions of angular point direction included angles, and the feature points which are matched incorrectly are removed. The specific method comprises the following steps:
the angular point direction of the characteristic point is obtained by a gray scale centroid method, namely, the angular point direction is represented by the vector direction formed by the angular point and the centroid by calculating the centroid of the gray scale of the pixel in the circular neighborhood of the angular point;
defining the angular point circular neighborhood moment as:the centroid of the angular point circular neighborhood is:and then the vector direction formed by the angular point and the mass center is the angular point direction of the feature point:；
wherein m ispqRepresenting p + q order moment, p and q are respectively a coefficient, I (x, y) represents the gray value of the pixel point (x, y) in the angular point circular neighborhood, (x, y) represents the coordinate of the pixel point in the angular point circular neighborhood, and m00Denotes the zero order moment, m10And m01Each represents an order moment;
Δθ i for the direction of an angular point, Delta, of a feature point in the remote sensing image to be registeredθ i ’The included angle of the angular point directions is delta for referring to the angular point directions of the corresponding characteristic points in the remote sensing imageθ i =Δθ i -Δθ i ’；
Then, the extracted ORB characteristic points are initially matched according to the following steps, and characteristic points which are wrongly matched are eliminated:
a. establishing an rBRIEF descriptor for the corner points extracted by the remote sensing image to be registered and the reference remote sensing image, setting the characteristic point set of the remote sensing image to be registered as { a1, a2, …, an1}, and setting the characteristic point set of the reference remote sensing image as { b1, b2, …, bn2 };
b. respectively comparing a descriptor of a characteristic point a1 of the remote sensing image to be registered with descriptors of all characteristic points b1, b2, … and bn2 of the reference remote sensing image by adopting a Brute-Force algorithm, calculating Hamming distances between the descriptor of a1 and the descriptors of b1, b2, … and bn2, selecting a point with the shortest Hamming distance from b1, b2, … and bn2, calculating a ratio of the shortest Hamming distance to the next shortest Hamming distance, setting a larger threshold value to be 0.8, if the ratio is smaller than the set larger threshold value, keeping the point with the shortest Hamming distance and the a1 as initial matching points, and calculating a direction corner point delta of the pointθ 1Otherwise, abandoning;
c. according to the method, the initial matching points of the characteristic points a2, … and an1 of the remote sensing image to be registered in the reference remote sensing image are sequentially solved, as shown in figure 4, and the corresponding angular point direction included angle delta is calculatedθ 2, …, Δθ n ；
d. Sequencing the initial paired points according to the ratio of the shortest Hamming distance to the second shortest Hamming distance from large to small, namely sequencing the matching quality of the feature points from good to bad, setting a smaller threshold value to be 0.5, and extracting the initial paired points of which the ratio of the shortest Hamming distance to the second shortest Hamming distance is smaller than the set smaller threshold value;
e. d, solving the optimal angular point direction included angle delta of the initial matching points extracted in the step d by a least square methodθ m ；
f. C, calculating the angle point direction included angles of all initial matching points calculated in the steps b and c, and taking the optimal angle point direction included angle deltaθ m For the constraint, the deviation Δθ m A range of initial paired point culling is shown in fig. 5.
Step S3: and carrying out parameter solution on the remote sensing image to be registered.
Considering that the transformation such as rotation, size and the like exists between the remote sensing image to be registered and the reference remote sensing image, determining a transformation matrix between the images to be H, wherein H is expressed as:
and (3) solving the optimal transformation parameters by using an RANSAC algorithm, wherein the process is as follows:
1) randomly extracting m samples from the matched feature point pairs, solving a transformation matrix H from the m samples, solving the homonymy point of the feature point in the remote sensing image to be registered in the reference remote sensing image according to the transformation matrix H, solving the distance between the homonymy point obtained by the transformation matrix H and the matching point obtained by Hamming distance matching, and taking the point with the distance smaller than a threshold value as an inner point;
2) repeating the steps for k times, and selecting a point set containing the maximum number of interior points;
3) and recalculating the transformation matrix H by using the samples in the selected point set so as to obtain the optimal transformation model which accords with most of the matching points.
Let the pixel points of the reference remote sensing image bef(x, y) The pixel point of the image to be registered isg(x’, y’) Assuming that the coordinates of a point on the reference remote sensing image are (x i , y i ) The coordinates of the corresponding points on the remote sensing image to be registered are (x i ’, y i ’) Then (1)x i , y i ) And (a)x i ’, y i ’) The affine transformation between is represented as:
in the formula,sis a scale factor, and is a function of,θis a rotation angle, ΔxAnd ΔyRespectively the translation amount of two coordinate axes;
after m (m is more than or equal to 4) feature points are obtained, an optimal transformation matrix is solved according to a RANSAC algorithm, namely, registration parameters are determineds、θ、Δx，Δy。
Step S4: and resampling the remote sensing image to be registered by adopting bilinear interpolation according to the registration parameters to finish image registration.
And (3) carrying out transformation such as scaling and rotation on the remote sensing image to be registered by using the solved transformation matrix H, resampling by adopting bilinear interpolation, fusing the two images in a mode of 0.5 multiplied by the reference image and 0.5 multiplied by the image to be registered, and finishing image registration as shown in figure 6.
The above are preferred embodiments of the present invention, and all changes made according to the technical scheme of the present invention that produce functional effects do not exceed the scope of the technical scheme of the present invention belong to the protection scope of the present invention.
Claims (5)
1. A method for rapidly registering remote sensing images is characterized by comprising the following steps:
step S1: respectively extracting ORB characteristic points of the remote sensing image to be registered and the reference remote sensing image;
step S2: performing initial matching on the extracted ORB characteristic points, and eliminating characteristic points which are in error matching from the initially matched characteristic points;
step S3: carrying out parameter solution on the remote sensing image to be registered;
step S4: and resampling the remote sensing image to be registered to finish image registration.
2. The method for rapidly registering remote sensing images according to claim 1, wherein in step S1, an image pyramid is established for the remote sensing image to be registered and the reference remote sensing image respectively, and ORB feature points are extracted for each layer of image pyramid.
3. The method for rapidly registering remote sensing images as claimed in claim 1, wherein in step S1, the method for extracting ORB feature points comprises: FAST angular point detection is firstly carried out, Harris angular point detection is carried out, the first N best points are selected, then non-maximum value inhibition is used for verifying angular points, and false edge points are removed, so that the feature points are distributed uniformly.
4. The method for rapidly registering remote sensing images according to claim 1, wherein in step S2, the extracted ORB feature points are initially matched by a hamming distance feature point matching method based on a dual threshold, and the feature points which are initially matched are rejected by using an angular point direction included angle as a constraint condition, and the method specifically comprises:
the angular point direction of the characteristic point is obtained by a gray scale centroid method, namely, the angular point direction is represented by the vector direction formed by the angular point and the centroid by calculating the centroid of the gray scale of the pixel in the circular neighborhood of the angular point;
defining the angular point circular neighborhood moment as:the centroid of the angular point circular neighborhood is:and then the vector direction formed by the angular point and the mass center is the angular point direction of the feature point:；
wherein m ispqRepresenting the order moment of p + q, I (x, y) representing the gray value of the pixel point (x, y) in the angular point circular neighborhood, (x, y) representing the coordinate of the pixel point in the angular point circular neighborhood, m00Denotes the zero order moment, m10And m01Each represents an order moment;
Δθ i for the direction of an angular point, Delta, of a feature point in the remote sensing image to be registeredθ i ’The included angle of the angular point directions is delta for referring to the angular point directions of the corresponding characteristic points in the remote sensing imageθ i =Δθ i -Δθ i ’；
Then, the extracted ORB characteristic points are initially matched according to the following steps, and characteristic points which are wrongly matched are eliminated:
a. establishing an rBRIEF descriptor for the corner points extracted by the remote sensing image to be registered and the reference remote sensing image, setting the characteristic point set of the remote sensing image to be registered as { a1, a2, …, an1}, and setting the characteristic point set of the reference remote sensing image as { b1, b2, …, bn2 };
b. respectively comparing the descriptor of the characteristic point a1 of the remote sensing image to be registered with the descriptors of all the characteristic points b1, b2, … and bn2 of the reference remote sensing image, calculating the Hamming distance between the descriptor of a1 and the descriptors of b1, b2, … and bn2, selecting the point with the shortest Hamming distance from the descriptors of b1, b2, … and bn2, calculating the ratio of the shortest Hamming distance to the next shortest Hamming distance, if the ratio is smaller than a set larger threshold value, keeping the point with the shortest Hamming distance and the a1 as initial matching points, and calculating the direction included angle delta of the angular point of the initial matching points and the angular point delta of the initial matching pointsθ 1Otherwise, abandoning;
c. according to the method, the initial matching points of the characteristic points a2, … and an1 of the remote sensing image to be registered in the reference remote sensing image are sequentially solved, and the corresponding angular point direction included angle delta is calculatedθ 2, …, Δθ n ；
d. Sequencing the initial matching points according to the ratio of the shortest Hamming distance to the second shortest Hamming distance from large to small, namely sequencing the matching quality of the feature points from good to poor, and extracting the initial matching points of which the ratio of the shortest Hamming distance to the second shortest Hamming distance is smaller than a set smaller threshold;
e. d, solving the optimal angular point direction included angle delta of the initial matching points extracted in the step d by a least square methodθ m ；
f. C, calculating the angle point direction included angles of all initial matching points calculated in the steps b and c, and taking the optimal angle point direction included angle deltaθ m For the constraint, the deviation Δθ m A range of initial matched point culling.
5. The method for rapidly registering remote sensing images according to claim 1, wherein in step S3, the method for performing parameter solution on the remote sensing images to be registered comprises:
let the pixel points of the reference remote sensing image bef(x, y) The pixel point of the image to be registered isg(x’, y’) Assuming that the coordinates of a point on the reference remote sensing image are (x i , y i ) The coordinates of the corresponding points on the remote sensing image to be registered are (x i ’, y i ’) Then (1)x i , y i ) And (a)x i ’, y i ’) The affine transformation between is represented as:
in the formula,sis a scale factor, and is a function of,θis a rotation angle, ΔxAnd ΔyRespectively the translation amount of two coordinate axes;
after m characteristic points are obtained, an optimal transformation matrix is solved according to a RANSAC algorithm, namely, registration parameters are determineds、θ、Δx，Δy；
And resampling the remote sensing image to be registered according to the registration parameters, namely finishing image registration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410434279.4A CN104167003B (en) | 2014-08-29 | 2014-08-29 | Method for fast registering remote-sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410434279.4A CN104167003B (en) | 2014-08-29 | 2014-08-29 | Method for fast registering remote-sensing image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104167003A true CN104167003A (en) | 2014-11-26 |
| CN104167003B CN104167003B (en) | 2017-01-18 |
Family
ID=51910794
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410434279.4A Active CN104167003B (en) | 2014-08-29 | 2014-08-29 | Method for fast registering remote-sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104167003B (en) |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105427263A (en) * | 2015-12-21 | 2016-03-23 | 努比亚技术有限公司 | Method and terminal for realizing image registering |
| CN105844616A (en) * | 2016-03-17 | 2016-08-10 | 湖南优象科技有限公司 | Binocular stereo matching algorithm under laser scattering spot auxiliary and apparatus thereof |
| TWI571805B (en) * | 2016-04-15 | 2017-02-21 | 元智大學 | Progressive image matching method and device based on hashing function |
| CN106651756A (en) * | 2016-11-16 | 2017-05-10 | 浙江工业大学 | Image registration method based on SIFT and authentication mechanism |
| CN106683046A (en) * | 2016-10-27 | 2017-05-17 | 山东省科学院情报研究所 | Real-time image splicing method for police unmanned aerial vehicle investigation and evidence obtaining |
| CN107240130A (en) * | 2017-06-06 | 2017-10-10 | 苍穹数码技术股份有限公司 | Remote Sensing Image Matching method, apparatus and system |
| CN107345812A (en) * | 2016-05-06 | 2017-11-14 | 湖北淦德智能消防科技有限公司 | A kind of image position method, device and mobile phone |
| CN107403127A (en) * | 2016-05-20 | 2017-11-28 | 重庆电信系统集成有限公司 | A kind of vehicle unloading state monitoring method based on image ORB features |
| CN107968916A (en) * | 2017-12-04 | 2018-04-27 | 国网山东省电力公司电力科学研究院 | A kind of fast video digital image stabilization method suitable for on-fixed scene |
| CN107993258A (en) * | 2017-11-23 | 2018-05-04 | 浙江大华技术股份有限公司 | A kind of method for registering images and device |
| CN108495089A (en) * | 2018-04-02 | 2018-09-04 | 北京京东尚科信息技术有限公司 | vehicle monitoring method, device, system and computer readable storage medium |
| CN108510530A (en) * | 2017-02-28 | 2018-09-07 | 深圳市朗驰欣创科技股份有限公司 | A kind of three-dimensional point cloud matching process and its system |
| CN108596867A (en) * | 2018-05-09 | 2018-09-28 | 五邑大学 | A kind of picture bearing calibration and system based on ORB algorithms |
| CN108805812A (en) * | 2018-06-04 | 2018-11-13 | 东北林业大学 | Multiple dimensioned constant ORB algorithms for image mosaic |
| CN109102534A (en) * | 2018-08-29 | 2018-12-28 | 长光卫星技术有限公司 | Optical remote sensing image registration method and system under the conditions of haze weather |
| CN109389630A (en) * | 2018-09-30 | 2019-02-26 | 北京精密机电控制设备研究所 | Visible images and the determination of Infrared Image Features point set, method for registering and device |
| CN109727279A (en) * | 2018-06-04 | 2019-05-07 | 南京师范大学 | A kind of autoegistration method of vector data and remote sensing image |
| CN110009670A (en) * | 2019-03-28 | 2019-07-12 | 上海交通大学 | The heterologous method for registering images described based on FAST feature extraction and PIIFD feature |
| CN110084784A (en) * | 2019-03-30 | 2019-08-02 | 天津大学 | Corner feature real-time detection and matching process on star |
| CN111445389A (en) * | 2020-02-24 | 2020-07-24 | 山东省科学院海洋仪器仪表研究所 | Wide-view-angle rapid splicing method for high-resolution images |
| CN111882594A (en) * | 2020-07-27 | 2020-11-03 | 北京环境特性研究所 | ORB feature point-based polarization image rapid registration method and device |
| CN112037193A (en) * | 2020-08-28 | 2020-12-04 | 长安大学 | Power line feature marking method and device |
| CN113192113A (en) * | 2021-04-30 | 2021-07-30 | 山东产研信息与人工智能融合研究院有限公司 | Binocular visual feature point matching method, system, medium and electronic device |
| CN113256653A (en) * | 2021-05-25 | 2021-08-13 | 南京信息工程大学 | High-rise ground object-oriented heterogeneous high-resolution remote sensing image registration method |
| CN113269817A (en) * | 2021-06-04 | 2021-08-17 | 北京中航世科电子技术有限公司 | Real-time remote sensing map splicing method and device combining spatial domain and frequency domain |
| CN113645443A (en) * | 2021-07-16 | 2021-11-12 | 南京理工大学 | FPGA-based surround video splicing display method and system |
| CN114004770A (en) * | 2022-01-04 | 2022-02-01 | 成都国星宇航科技有限公司 | Method and device for accurately correcting satellite space-time diagram and storage medium |
| US11354883B2 (en) | 2019-12-30 | 2022-06-07 | Sensetime International Pte. Ltd. | Image processing method and apparatus, and electronic device |
| CN116152532A (en) * | 2023-04-14 | 2023-05-23 | 中国地质大学（武汉） | Remote sensing image feature extraction and matching method and device and electronic equipment |
| CN116434072A (en) * | 2023-06-12 | 2023-07-14 | 山东省国土空间生态修复中心(山东省地质灾害防治技术指导中心、山东省土地储备中心) | Geological disaster early identification method and device based on multi-source data |
-
2014
- 2014-08-29 CN CN201410434279.4A patent/CN104167003B/en active Active
Non-Patent Citations (1)
| Title |
|---|
| 张云生等: "基于改进 OＲB 算法的遥感图像自动配准方法", 《国土资源遥感》 * |
Cited By (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105427263A (en) * | 2015-12-21 | 2016-03-23 | 努比亚技术有限公司 | Method and terminal for realizing image registering |
| CN105844616A (en) * | 2016-03-17 | 2016-08-10 | 湖南优象科技有限公司 | Binocular stereo matching algorithm under laser scattering spot auxiliary and apparatus thereof |
| TWI571805B (en) * | 2016-04-15 | 2017-02-21 | 元智大學 | Progressive image matching method and device based on hashing function |
| CN107345812A (en) * | 2016-05-06 | 2017-11-14 | 湖北淦德智能消防科技有限公司 | A kind of image position method, device and mobile phone |
| CN107403127A (en) * | 2016-05-20 | 2017-11-28 | 重庆电信系统集成有限公司 | A kind of vehicle unloading state monitoring method based on image ORB features |
| CN106683046A (en) * | 2016-10-27 | 2017-05-17 | 山东省科学院情报研究所 | Real-time image splicing method for police unmanned aerial vehicle investigation and evidence obtaining |
| CN106683046B (en) * | 2016-10-27 | 2020-07-28 | 山东省科学院情报研究所 | Image real-time splicing method for police unmanned aerial vehicle reconnaissance and evidence obtaining |
| CN106651756A (en) * | 2016-11-16 | 2017-05-10 | 浙江工业大学 | Image registration method based on SIFT and authentication mechanism |
| CN106651756B (en) * | 2016-11-16 | 2020-05-01 | 浙江工业大学 | Image registration method based on SIFT and verification mechanism |
| CN108510530A (en) * | 2017-02-28 | 2018-09-07 | 深圳市朗驰欣创科技股份有限公司 | A kind of three-dimensional point cloud matching process and its system |
| CN107240130A (en) * | 2017-06-06 | 2017-10-10 | 苍穹数码技术股份有限公司 | Remote Sensing Image Matching method, apparatus and system |
| CN107993258A (en) * | 2017-11-23 | 2018-05-04 | 浙江大华技术股份有限公司 | A kind of method for registering images and device |
| CN107993258B (en) * | 2017-11-23 | 2021-02-02 | 浙江大华技术股份有限公司 | Image registration method and device |
| CN107968916A (en) * | 2017-12-04 | 2018-04-27 | 国网山东省电力公司电力科学研究院 | A kind of fast video digital image stabilization method suitable for on-fixed scene |
| CN108495089A (en) * | 2018-04-02 | 2018-09-04 | 北京京东尚科信息技术有限公司 | vehicle monitoring method, device, system and computer readable storage medium |
| CN108596867A (en) * | 2018-05-09 | 2018-09-28 | 五邑大学 | A kind of picture bearing calibration and system based on ORB algorithms |
| CN109727279A (en) * | 2018-06-04 | 2019-05-07 | 南京师范大学 | A kind of autoegistration method of vector data and remote sensing image |
| CN108805812A (en) * | 2018-06-04 | 2018-11-13 | 东北林业大学 | Multiple dimensioned constant ORB algorithms for image mosaic |
| CN109727279B (en) * | 2018-06-04 | 2022-07-29 | 南京师范大学 | Automatic registration method of vector data and remote sensing image |
| CN109102534A (en) * | 2018-08-29 | 2018-12-28 | 长光卫星技术有限公司 | Optical remote sensing image registration method and system under the conditions of haze weather |
| CN109102534B (en) * | 2018-08-29 | 2020-09-01 | 长光卫星技术有限公司 | Optical remote sensing image registration method and system under haze weather condition |
| CN109389630A (en) * | 2018-09-30 | 2019-02-26 | 北京精密机电控制设备研究所 | Visible images and the determination of Infrared Image Features point set, method for registering and device |
| CN109389630B (en) * | 2018-09-30 | 2020-10-23 | 北京精密机电控制设备研究所 | Method and device for determining and registering feature point set of visible light image and infrared image |
| CN110009670A (en) * | 2019-03-28 | 2019-07-12 | 上海交通大学 | The heterologous method for registering images described based on FAST feature extraction and PIIFD feature |
| CN110084784A (en) * | 2019-03-30 | 2019-08-02 | 天津大学 | Corner feature real-time detection and matching process on star |
| US11354883B2 (en) | 2019-12-30 | 2022-06-07 | Sensetime International Pte. Ltd. | Image processing method and apparatus, and electronic device |
| CN111445389A (en) * | 2020-02-24 | 2020-07-24 | 山东省科学院海洋仪器仪表研究所 | Wide-view-angle rapid splicing method for high-resolution images |
| CN111882594A (en) * | 2020-07-27 | 2020-11-03 | 北京环境特性研究所 | ORB feature point-based polarization image rapid registration method and device |
| CN112037193A (en) * | 2020-08-28 | 2020-12-04 | 长安大学 | Power line feature marking method and device |
| CN113192113A (en) * | 2021-04-30 | 2021-07-30 | 山东产研信息与人工智能融合研究院有限公司 | Binocular visual feature point matching method, system, medium and electronic device |
| CN113192113B (en) * | 2021-04-30 | 2022-12-23 | 山东产研信息与人工智能融合研究院有限公司 | Binocular visual feature point matching method, system, medium and electronic device |
| CN113256653A (en) * | 2021-05-25 | 2021-08-13 | 南京信息工程大学 | High-rise ground object-oriented heterogeneous high-resolution remote sensing image registration method |
| CN113256653B (en) * | 2021-05-25 | 2023-05-09 | 南京信息工程大学 | Heterogeneous high-resolution remote sensing image registration method for high-rise ground object |
| CN113269817A (en) * | 2021-06-04 | 2021-08-17 | 北京中航世科电子技术有限公司 | Real-time remote sensing map splicing method and device combining spatial domain and frequency domain |
| CN113645443A (en) * | 2021-07-16 | 2021-11-12 | 南京理工大学 | FPGA-based surround video splicing display method and system |
| CN113645443B (en) * | 2021-07-16 | 2022-05-13 | 南京理工大学 | FPGA-based surround video splicing display method and system |
| CN114004770B (en) * | 2022-01-04 | 2022-04-26 | 成都国星宇航科技有限公司 | Method and device for accurately correcting satellite space-time diagram and storage medium |
| CN114004770A (en) * | 2022-01-04 | 2022-02-01 | 成都国星宇航科技有限公司 | Method and device for accurately correcting satellite space-time diagram and storage medium |
| CN116152532A (en) * | 2023-04-14 | 2023-05-23 | 中国地质大学（武汉） | Remote sensing image feature extraction and matching method and device and electronic equipment |
| CN116434072A (en) * | 2023-06-12 | 2023-07-14 | 山东省国土空间生态修复中心(山东省地质灾害防治技术指导中心、山东省土地储备中心) | Geological disaster early identification method and device based on multi-source data |
| CN116434072B (en) * | 2023-06-12 | 2023-08-18 | 山东省国土空间生态修复中心(山东省地质灾害防治技术指导中心、山东省土地储备中心) | Geological disaster early identification method and device based on multi-source data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104167003B (en) | 2017-01-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104167003B (en) | Method for fast registering remote-sensing image | |
| CN110332887B (en) | Monocular vision pose measurement system and method based on characteristic cursor points | |
| CN103236064B (en) | A kind of some cloud autoegistration method based on normal vector | |
| CN109903313B (en) | Real-time pose tracking method based on target three-dimensional model | |
| CN110992263B (en) | Image stitching method and system | |
| CN105160686B (en) | A kind of low latitude various visual angles Remote Sensing Images Matching Method based on improvement SIFT operators | |
| CN111145232A (en) | Three-dimensional point cloud automatic registration method based on characteristic information change degree | |
| CN109241901B (en) | A kind of detection and recognition methods to the three-dimensional point cloud with hole | |
| CN106355577A (en) | Method and system for quickly matching images on basis of feature states and global consistency | |
| CN108830888B (en) | Coarse matching method based on improved multi-scale covariance matrix characteristic descriptor | |
| CN102800099B (en) | Multi-feature multi-level visible light and high-spectrum image high-precision registering method | |
| CN107818598B (en) | Three-dimensional point cloud map fusion method based on visual correction | |
| CN108550166B (en) | Spatial target image matching method | |
| CN107862708A (en) | A kind of SAR and visible light image registration method | |
| CN108269274B (en) | Image registration method based on Fourier transform and Hough transform | |
| CN114331879A (en) | Visible light and infrared image registration method for equalized second-order gradient histogram descriptor | |
| CN111199558A (en) | Image matching method based on deep learning | |
| CN113095385A (en) | Multimode image matching method based on global and local feature description | |
| CN107808165B (en) | Infrared image matching method based on SUSAN corner detection | |
| CN103077528A (en) | Rapid image matching method based on DCCD (Digital Current Coupling)-Laplace and SIFT (Scale Invariant Feature Transform) descriptors | |
| CN112465876A (en) | Stereo matching method and equipment | |
| CN110232694B (en) | Infrared polarization thermal image threshold segmentation method | |
| CN112418250B (en) | Optimized matching method for complex 3D point cloud | |
| CN102679871B (en) | Rapid detection method of sub-pixel precision industrial object | |
| CN104484647A (en) | High-resolution remote sensing image cloud height detection method |
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 |