CN113160071A - Satellite image automatic geometric correction method, system, medium and terminal equipment - Google Patents
Satellite image automatic geometric correction method, system, medium and terminal equipment Download PDFInfo
- Publication number
- CN113160071A CN113160071A CN202110266590.2A CN202110266590A CN113160071A CN 113160071 A CN113160071 A CN 113160071A CN 202110266590 A CN202110266590 A CN 202110266590A CN 113160071 A CN113160071 A CN 113160071A
- Authority
- CN
- China
- Prior art keywords
- image
- target satellite
- modis
- images
- correction
- 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
- 238000012937 correction Methods 0.000 title claims abstract description 164
- 238000000034 method Methods 0.000 title claims abstract description 76
- 238000012216 screening Methods 0.000 claims abstract description 48
- 238000012545 processing Methods 0.000 claims abstract description 16
- 238000011156 evaluation Methods 0.000 claims abstract description 14
- 238000004590 computer program Methods 0.000 claims description 18
- 108091092919 Minisatellite Proteins 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000000605 extraction Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims 1
- 238000003702 image correction Methods 0.000 abstract description 6
- 230000007547 defect Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 3
- 108091092878 Microsatellite Proteins 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000002310 reflectometry Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10032—Satellite or aerial image; Remote sensing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20132—Image cropping
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method, a device, a storage medium and terminal equipment for automatically geometrically correcting domestic polar region small satellite images, which are used for automatically and preferentially screening homologous sensor reference images for automatically geometrically correcting target satellite images; dividing the whole scene image and the image into four parts and nine parts, carrying out image enhancement processing, and extracting homonymous point pairs of the target satellite image and the reference image to obtain a geographic coordinate file for geometric correction; and carrying out correction precision evaluation on the image corrected by the target satellite by adopting different methods to screen out an optimal correction scheme, and correcting the image of the target satellite according to the optimal correction scheme. The method can realize automatic preferential screening of the homologous sensor registration reference images, highlight the detail characteristics of polar images by an image enhancement method to increase the selection of the same-name points, and screen out an optimal geometric correction scheme according to the image correction precision so as to overcome the defect of low geometric positioning precision of the polar small satellite in China.
Description
Technical Field
The invention relates to the technical field of remote sensing image processing, in particular to a satellite image automatic geometric correction method, a satellite image automatic geometric correction system, a storage medium and terminal equipment.
Background
The existing image registration aiming at the homologous sensor is mainly carried out by adopting a characteristic-based method, but in the front of mass data registration, the registration reference image of the homologous sensor and the selection of homologous points are difficult to automatically and preferentially screen. Particularly, in polar region images, due to the fact that the ice and snow reflectivity is high, the texture of the surface of the ice cover is rare, and the feature points extracted by the scale-invariant feature transformation algorithm on the surface of the ice cover are few, so that difficulty is brought to geometric correction of the polar region images.
Disclosure of Invention
The technical problem to be solved by the embodiments of the present invention is to provide a method, a system, a storage medium and a terminal device for automatically and preferentially screening a homologous sensor registration reference image, and to highlight detail features of polar images by an image enhancement method to increase selection of homologous points, and to screen out an optimal geometric correction scheme according to image correction accuracy, so as to overcome the defect of low geometric positioning accuracy of a minisatellite in the polar region in China.
In order to solve the above technical problem, an embodiment of the present invention provides an automatic geometric correction method for satellite images, which is applied to a polar region minisatellite, and the method includes:
preferentially screening a reference image for the automatic geometric correction of the target satellite image from three aspects of image acquisition time distance, space coverage degree and number of homonymous point pairs to obtain an optimal reference image for the geometric correction of the target satellite image;
extracting homonymous points of the target satellite image and the MODIS image by adopting a preset scheme to obtain a geographic coordinate file for geometric correction of each scheme; the preset scheme comprises the steps of extracting homonymous points from a whole scene image, dividing the image into four parts and nine parts, carrying out image enhancement processing, and then extracting homonymous points;
and carrying out correction precision evaluation on the image corrected by the target satellite image by adopting a preset scheme so as to screen out an optimal correction scheme, and correcting the target satellite image according to the optimal correction scheme.
Further, the method preferentially screens the MODIS images for the automatic geometric correction of the target satellite image in three aspects of image acquisition time distance, spatial coverage degree and number of homonymous point pairs to obtain an optimal reference image for the geometric correction of the target satellite image, and specifically comprises the following steps:
obtaining MODIS images of a target satellite image which is cloudy in the day in batches, and creating a first data index table of the target satellite image and the MODIS images;
preprocessing a target satellite image and an MODIS image, and obtaining the MODIS image which is consistent with the target satellite image in geographic range by adopting consistent south pole projection and 250m resolution;
preliminarily screening the MODIS images, screening images with the effective data range covering more than 80% of the target images, and creating a second data index table of the target satellite images and the MODIS images;
and carrying out secondary screening on the MODIS images, extracting homonymous points of a target satellite and an MODIS image pair according to the second data index table by a feature matching method, sorting and preferentially selecting corresponding MODIS images according to the number of homonymous point pairs, and if the multi-scene MODIS data still have the same number of homonymous point pairs, taking the MODIS image close to the shooting time of the target satellite image as a final registration reference image and obtaining a third data index table for geometric correction.
Further, extracting the homonymous points of the target satellite image and the MODIS image by adopting a preset scheme to obtain a geographic coordinate file of each scheme for geometric correction, specifically:
the first scheme is as follows: extracting homonymous point pairs from a target satellite panoramic image and a corresponding MODIS panoramic image through a scale invariant feature transform algorithm, sorting according to Euclidean distances of the homonymous point pairs, removing 10% of points with the largest Euclidean distances, and outputting real geographic coordinate files of the rest homonymous point pairs as first corrected geographic coordinate files for correcting the target satellite image;
scheme II: dividing a target image into four parts according to rules, cutting the MODIS image corresponding to 40 pixels extending from each part, performing piecewise linear stretching enhancement processing on each image pair, extracting dotted pairs through a scale invariant feature transform algorithm, removing points with the maximum Euclidean distance of 30%, outputting real geographic coordinate files of the rest dotted points, combining the coordinate file and a first corrected geographic coordinate file, and removing repeated point pairs to serve as a second corrected geographic coordinate file for correcting the target satellite image;
the third scheme is as follows: and dividing the target image into nine parts according to rules, obtaining real geographic coordinate files of the other same-name points after the points with the maximum Euclidean distance of 30% are removed by the same method as the scheme II, combining the coordinate files with the first corrected geographic coordinate file and the second corrected geographic coordinate file, and removing repeated point pairs to serve as a third corrected geographic coordinate file for correcting the target satellite image.
Further, the image corrected by different methods is subjected to correction precision evaluation to screen out an optimal correction scheme, and the target satellite image is corrected according to the optimal correction scheme, specifically:
respectively using three corrected geographic coordinate files obtained by the three schemes for geometric correction of the target image by using a quadratic polynomial method to obtain target satellite images corrected by different schemes;
extracting the homonymous points of the three corrected target satellite images and the corresponding MODIS images by using a scale invariant feature transform algorithm, calculating Euclidean distances between the point pairs, evaluating correction precision after geometric correction of different schemes, and screening out an optimal correction scheme;
and performing geometric correction on the target satellite image by using the optimal correction scheme.
In order to solve the above technical problem, an embodiment of the present invention further provides an automatic geometric correction system for satellite images, including:
the screening module is used for preferentially screening the reference image for the automatic geometric correction of the target satellite image from three aspects of image acquisition time, space coverage degree and number of homonymous point pairs so as to obtain the optimal reference image for the geometric correction of the target satellite image;
the extraction module is used for extracting the homonymous points of the target satellite image and the MODIS image by adopting a preset scheme so as to obtain a geographic coordinate file of each scheme for geometric correction; the preset scheme comprises the steps of extracting homonymous points from a whole scene image, dividing the image into four parts and nine parts, carrying out image enhancement processing, and then extracting homonymous points;
and the correction module is used for carrying out correction precision evaluation on the image corrected by the target satellite image by adopting a preset scheme so as to screen out an optimal correction scheme and correcting the target satellite image according to the optimal correction scheme.
Further, the screening module, in particular for,
obtaining MODIS images of a target satellite image which is cloudy in the day in batches, and creating a first data index table of the target satellite image and the MODIS images;
preprocessing a target satellite image and an MODIS image, and obtaining the MODIS image which is consistent with the target satellite image in geographic range by adopting consistent south pole projection and 250m resolution;
preliminarily screening the MODIS images, screening images with the effective data range covering more than 80% of the target images, and creating a second data index table of the target satellite images and the MODIS images;
and carrying out secondary screening on the MODIS images, extracting homonymous points of a target satellite and an MODIS image pair according to the second data index table by a feature matching method, sorting and preferentially selecting corresponding MODIS images according to the number of homonymous point pairs, and if the multi-scene MODIS data still have the same number of homonymous point pairs, taking the MODIS image close to the shooting time of the target satellite image as a final registration reference image and obtaining a third data index table for geometric correction.
Further, the extraction module, in particular for,
the first scheme is as follows: extracting homonymous point pairs from a target satellite panoramic image and a corresponding MODIS panoramic image through a scale invariant feature transform algorithm, sorting according to Euclidean distances of the homonymous point pairs, removing 10% of points with the largest Euclidean distances, and outputting real geographic coordinate files of the rest homonymous point pairs as first corrected geographic coordinate files for correcting the target satellite image;
scheme II: dividing a target image into four parts according to rules, cutting the MODIS image corresponding to 40 pixels extending from each part, performing piecewise linear stretching enhancement processing on each image pair, extracting dotted pairs through a scale invariant feature transform algorithm, removing points with the maximum Euclidean distance of 30%, outputting real geographic coordinate files of the rest dotted points, combining the coordinate file and a first corrected geographic coordinate file, and removing repeated point pairs to serve as a second corrected geographic coordinate file for correcting the target satellite image;
the third scheme is as follows: and dividing the target image into nine parts according to rules, obtaining real geographic coordinate files of the other same-name points after the points with the maximum Euclidean distance of 30% are removed by the same method as the scheme II, combining the coordinate files with the first corrected geographic coordinate file and the second corrected geographic coordinate file, and removing repeated point pairs to serve as a third corrected geographic coordinate file for correcting the target satellite image.
Further, the correction module, in particular for,
respectively using three corrected geographic coordinate files obtained by the three schemes for geometric correction of the target image by using a quadratic polynomial method to obtain target satellite images corrected by different schemes;
extracting the homonymous points of the three corrected target satellite images and the corresponding MODIS images by using a scale invariant feature transform algorithm, calculating Euclidean distances between the point pairs, evaluating correction precision after geometric correction of different schemes, and screening out an optimal correction scheme;
and performing geometric correction on the target satellite image by using the optimal correction scheme.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program; when the computer program runs, the computer program controls the equipment where the computer readable storage medium is located to execute the satellite image automatic geometric correction method.
The embodiment of the invention also provides terminal equipment which comprises a processor, a memory and a computer program which is stored in the memory and is configured to be executed by the processor, wherein the processor realizes the satellite image automatic geometric correction method when executing the computer program.
Compared with the prior art, the embodiment of the invention provides an automatic geometric correction method for satellite images, which comprises the steps of firstly, preferentially screening a reference MODIS image for automatic geometric correction of a target satellite image from three aspects of image acquisition time distance, space coverage degree and number of homonymous point pairs to obtain an optimal reference image for geometric correction of the target satellite image; then, extracting homonymous points of the target satellite image and the MODIS image by adopting different methods, wherein the homonymous points are extracted by extracting the homonymous points from the whole scene image, dividing the image into four parts and nine parts, and then extracting homonymous points to obtain a geographic coordinate file for geometric correction of each scheme; and finally, carrying out correction precision evaluation on the image corrected by the target satellite image by adopting different methods to screen out an optimal correction scheme, and correcting the target satellite image according to the optimal correction scheme. Compared with the prior art, the method can realize automatic preferred screening of the homologous sensor registration reference image, highlight the detail characteristics of the polar image by an image enhancement method to increase the selection of the homologous points, and screen out the optimal geometric correction scheme according to the image correction precision so as to overcome the defect of low geometric positioning precision of the existing polar minisatellite.
Drawings
FIG. 1 is a flow chart of an automated geometric correction method for satellite images according to the present invention;
FIG. 2 is a data flow diagram of an automated geometric correction method for satellite images according to the present invention;
FIG. 3 is a diagram illustrating the effect of different image enhancements in an automated geometric correction method for satellite images according to the present invention;
FIG. 4 is a diagram illustrating an effect of an automated geometric correction method for satellite images according to the present invention;
FIG. 5 is a block diagram illustrating an automated geometric correction method for satellite images according to the present invention;
fig. 6 is a block diagram of a terminal device according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without any inventive step, are within the scope of the present invention.
It should be noted that, the step numbers in the text are only for convenience of explanation of the specific embodiments, and do not serve to limit the execution sequence of the steps. The method provided by the embodiment can be executed by the relevant server, and the server is taken as an example for explanation below.
As shown in fig. 1 to 4, an embodiment of the invention provides an automatic geometric correction method for satellite images, which is applied to a minisatellite in polar region, and the method includes steps S11 to S14:
step S11, preferentially screening the reference images for the automatic geometric correction of the target satellite image from the three aspects of the image acquisition time, the spatial coverage degree and the number of the homonymous point pairs to obtain the optimal reference image for the geometric correction of the target satellite image.
Specifically, MODIS images with few clouds of a target satellite image in the same day are obtained in batches, and a first data index table of the target satellite image and the MODIS images is created; preprocessing the target satellite image and the MODIS image, and obtaining the MODIS image which is consistent with the geographic range of the target satellite image by image cutting by adopting consistent projection (south pole projection) and resolution (250 m); preliminarily screening the MODIS images, screening images of which the effective data range covers more than a certain target value of the target satellite image, and creating a second data index table of the target satellite image and the MODIS images; and carrying out secondary screening on the MODIS images, extracting homonymous points of each target satellite and an MODIS image pair according to the second data index table by a feature matching method, sorting and preferentially selecting corresponding MODIS images according to the number of homonymous point pairs, and if multi-scene MODIS data still have the same number of homonymous point pairs, taking the MODIS images close to the shooting time of the target satellite as final registration reference images and obtaining a third data index table for geometric correction.
Further, taking the minisatellite 'Kyoto I' (research code: BNU-1) satellite in the polar region of China as an example, the MODIS image (American MODIS reflectivity data) of the same day corresponding to the image corresponding to the BNU-1 satellite is downloaded in batch, and a first data index table of the two is established. BNU-1 data defines the south pole projection (set resolution 250m) (as in MODIS data); extracting single wave band of radiance from the MODIS image, recovering geographic coordinates, outputting GeoTIFF, defining a south pole projection (the resolution is set to be 250m), and cutting the MODIS image (the BNU-1 image with the resolution of 250m extends 40 pixels (namely 10km) to cut the corresponding MODIS image). And when the cut MODIS effective data range/corresponding BNU-1 data range is more than or equal to 0.8, creating a second data index table of the BNU-1 satellite image and the MODIS image. And extracting homonymous point pairs of the BNU-1 satellite image and the corresponding MODIS image by using a Scale Invariant Feature Transform (SIFT) matching method according to the second data index table. And (3) sorting and preferentially selecting corresponding MODIS images according to the number of homonymous point pairs, and if the multi-scene MODIS data have the same number of homonymous point pairs, taking the MODIS data close to the BNU-1 shooting time as a final registration reference image to obtain a third data index table finally used for registration.
Step S12, extracting the homonymous points of the target satellite image and the MODIS image by adopting a preset scheme to obtain a geographic coordinate file for geometric correction of each scheme; the preset scheme comprises the steps of extracting homonymous points from the whole scene image, dividing the image into four parts and nine parts, carrying out image enhancement processing, and then extracting homonymous points.
Wherein, the first scheme is as follows: extracting the coordinates of the homonymous points of the homonymous point pairs by an integer point image method to obtain a corrected geographic coordinate file, which specifically comprises the following steps: extracting homonymous point pairs from the target satellite panoramic image and the correspondingly cut MODIS panoramic image through a SIFT algorithm, sorting according to the Euclidean distances of the homonymous point pairs, removing the points with the maximum 10% of the Euclidean distances, and outputting real geographic coordinate files of the rest homonymous point pairs as first correction geographic coordinate files for correcting the automatic geometric correction of the target satellite image.
Specifically, the BNU-1 panoramic image and the correspondingly cut MODIS panoramic image (according to a third data index table) are used for extracting homonymous point pairs by utilizing an SIFT algorithm, sorting is carried out according to the Euclidean distances of the homonymous point pairs, the point with the maximum Euclidean distance of 10% is removed, and the real geographic coordinate files of the rest homonymous point pairs are output as a first correction geographic coordinate file for correcting the BNU-1 satellite image.
Wherein, the scheme II: the method comprises the steps of regularly dividing a target satellite image into four parts (2 x 2 to 4), cutting each part into 40 pixels extending outwards to form a corresponding MODIS image, conducting piecewise linear stretching enhancement processing on each image pair, extracting same-name point pairs through an SIFT algorithm, eliminating points with the maximum Euclidean distance of 30%, outputting real geographic coordinate files of the rest same-name points, combining the coordinate file and a first corrected geographic coordinate file, and eliminating repeated point pairs to serve as a second corrected geographic coordinate file for correcting the BNU-1 satellite image.
Specifically, referring to fig. 3, the BNU-1 image is regularly clipped into 4 portions (2 × 2 — 4), and the four portions are extended by 40 pixels to clip the corresponding MODIS data. And respectively carrying out image piecewise linear stretching enhancement on the four groups of image pairs, and then extracting the same-name point pairs by using an SIFT algorithm. The comparison of the homonymous point pairs extracted by different stretching methods in fig. 3 is: (a) no stretching; (b) linear stretching; (c) stretching in a piecewise linear way; (d) gaussian stretching; (e) performing histogram equalization stretching; (f) root mean square stretching (comparing results of six stretching modes of no stretching, linear stretching, piecewise linear stretching, Gaussian stretching, histogram equalization and root mean square, more homonymous points are found out by the piecewise linear stretching mode), points with the maximum Euclidean distance of 30% are removed, and real geographic coordinate files of the rest homonymous points are output. And combining the coordinate file with the first corrected geographic coordinate file, and then removing the repeated point pairs to serve as a second corrected geographic coordinate file for correcting the BNU-1 satellite image. It can be understood that the characteristic of the ice and snow in the polar region can be highlighted by acquiring the homonymous points after image enhancement, so that more homonymous points can be found.
Wherein, the third scheme is as follows: and (3) dividing the target satellite image and the MODIS image into nine parts according to rules (3 x 3 is 9, then extracting homonymy point pairs, removing points with the maximum Euclidean distance of 30%, and outputting real geographic coordinate files of the rest homonymy points (the method is the same as the scheme II).
Specifically, the BNU-1 satellite image and the MODIS image are uniformly divided into 9 parts for piecewise linear enhancement, then homonymous point pairs are extracted, and after 30% of homonymous point pairs are removed, a coordinate file is output (the method is the same as the scheme II). And after the first corrected geographic coordinate file and the second corrected geographic coordinate file are combined, the repeated point pairs are removed and used as a third corrected geographic coordinate file for correcting the BNU-1 satellite image.
And step S13, performing correction precision evaluation on the image corrected by the target satellite image by adopting different schemes to screen out an optimal correction scheme, and correcting the target satellite image according to the optimal correction scheme.
Specifically, three corrected geographic coordinate files obtained by the three schemes are respectively used for geometric correction of the target satellite image by utilizing a quadratic polynomial method, so that the target satellite image corrected by the different schemes is obtained; extracting the homonymous points of the corrected BNU-1 satellite images (three types) and the corresponding MODIS images again by utilizing an SIFT algorithm, calculating the Euclidean distance between the point pairs, evaluating the correction precision after geometric correction of different schemes, and screening out the optimal correction scheme; and carrying out geometric correction on the BNU-1 satellite image by using an optimal correction scheme.
Further, three corrected geographic coordinate files obtained by the three schemes are respectively used for geometric correction of the BNU-1 satellite image by utilizing a quadratic polynomial method to obtain corrected images of different schemes; and (3) extracting the homonymous points of the corrected BNU-1 satellite images (three types) and the corresponding MODIS images again by utilizing an SIFT algorithm, calculating the Euclidean distance between the point pairs, evaluating the correction precision (table 1) of different schemes after geometric correction, and screening out the optimal correction scheme.
TABLE 1 evaluation of the quantity of homonymous points and correction accuracy for BNU-1 satellite image automatic geometric correction by three schemes
(for example, the image taken in 2019 by BNU-1 satellite, 12 month, 18 days 09:04: 51)
The automatic geometric correction technology suitable for polar microsatellite images is applied to BNU-1 satellite images for geometric correction, corresponding MODIS data are respectively superposed on the images before and after correction at the lower layer, and the effects before and after correction are compared (taking the image shot by BNU-1 satellite 2019 at 12/18/09: 04:51 as an example), and the result is shown in FIG. 4.
In FIG. 4, FIG. 4a is a BNU-10 level image, and FIG. 4b is a corrected image using an automatic geometric correction technique suitable for domestic minisatellite images. Fig. 4c, e, g are respective enlargements of the detail in fig. 4a, respectively sea ice boundary, ice cover and cloudy portion. Fig. 4d, f, h are enlarged views of the corresponding positions of the corrected images using the technique of the present invention. From the results, the automatic geometric correction technology suitable for the domestic polar region minisatellite image has obvious advantages in the following aspects, firstly, the technology can automatically and preferentially screen the registration reference image of the homologous sensor (the optimal scene can be selected from dozens of scenes of data every day), and the manual intervention brought by the aspects of image data range, shooting time, data quality and the like due to large satellite data volume is avoided. Secondly, the image is subjected to subsection enhancement processing by the technology, the detail features of the polar image are highlighted, and more homonymy point pairs can be extracted more and more uniformly by the SIFT operator on the polar image. Thirdly, the technology can select a correction scheme according to different image preferences to achieve optimal correction precision. Fourthly, the geometric accuracy of the corrected images is remarkably improved (since 6221.77m is improved to 243.48m), a series of high-accuracy registered data products can be automatically produced, and effective support and guarantee are provided for improving and promoting the application of the minimal satellite data.
The embodiment of the invention provides an automatic geometric correction method for satellite images, which comprises the steps of firstly, preferentially screening a reference MODIS image for automatic geometric correction of a target satellite image from three aspects of time acquisition, time proximity, space coverage degree and number of homonymous point pairs to obtain an optimal reference image for geometric correction of the target satellite image; then, extracting homonymous points of the target satellite image and the MODIS image by adopting different methods, wherein the homonymous points are extracted by extracting the homonymous points from the whole scene image, dividing the image into four parts and nine parts, and then extracting homonymous points to obtain a geographic coordinate file for geometric correction of each scheme; and finally, carrying out correction precision evaluation on the image corrected by the target satellite image by adopting different methods to screen out an optimal correction scheme, and correcting the target satellite image according to the optimal correction scheme. Compared with the prior art, the method can realize automatic preferred screening of the homologous sensor registration reference image, highlight the detail characteristics of the polar image by an image enhancement method to increase the selection of the homonymy point, and screen out the optimal geometric correction scheme according to the image correction precision, so as to overcome the defect of low geometric positioning precision of the polar microsatellite in China and meet the actual application requirement.
As shown in fig. 5, it is a block diagram of an automatic geometry correction system for satellite images according to the present invention, the system includes:
the screening module 21 is configured to preferentially screen a reference image for automatic geometric correction of the target satellite image from three aspects, namely, a time of image acquisition, a spatial coverage degree, and a number of homonymous point pairs, so as to obtain an optimal reference image for geometric correction of the target satellite image.
Further, the screening module 21 is specifically configured to,
obtaining MODIS images of a target satellite image which is cloudy in the day in batches, and creating a first data index table of the target satellite image and the MODIS images;
preprocessing a target satellite image and an MODIS image, and obtaining the MODIS image which is consistent with the target satellite image in geographic range by adopting consistent south pole projection and 250m resolution;
preliminarily screening the MODIS images, screening images with the effective data range covering more than 80% of the target images, and creating a second data index table of the target satellite images and the MODIS images;
and carrying out secondary screening on the MODIS images, extracting homonymous points of a target satellite and an MODIS image pair according to the second data index table by a feature matching method, sorting and preferentially selecting corresponding MODIS images according to the number of homonymous point pairs, and if the multi-scene MODIS data still have the same number of homonymous point pairs, taking the MODIS image close to the shooting time of the target satellite image as a final registration reference image and obtaining a third data index table for geometric correction.
The extraction module 22 is configured to extract the homonymous points of the target satellite image and the MODIS image by using a preset scheme to obtain a geographic coordinate file for geometric correction of each scheme; the preset scheme comprises the steps of extracting homonymous points from the whole scene image, dividing the image into four parts and nine parts, carrying out image enhancement processing, and then extracting homonymous points.
Further, the extraction module 22 is, in particular,
the first scheme is as follows: extracting homonymous point pairs from a target satellite panoramic image and a corresponding MODIS panoramic image through a scale invariant feature transform algorithm, sorting according to Euclidean distances of the homonymous point pairs, removing 10% of points with the largest Euclidean distances, and outputting real geographic coordinate files of the rest homonymous point pairs as first corrected geographic coordinate files for correcting the target satellite image;
scheme II: dividing a target image into four parts according to rules, cutting the MODIS image corresponding to 40 pixels extending from each part, performing piecewise linear stretching enhancement processing on each image pair, extracting dotted pairs through a scale invariant feature transform algorithm, removing points with the maximum Euclidean distance of 30%, outputting real geographic coordinate files of the rest dotted points, combining the coordinate file and a first corrected geographic coordinate file, and removing repeated point pairs to serve as a second corrected geographic coordinate file for correcting the target satellite image;
the third scheme is as follows: and dividing the target image into nine parts according to rules, obtaining real geographic coordinate files of the other same-name points after the points with the maximum Euclidean distance of 30% are removed by the same method as the scheme II, combining the coordinate files with the first corrected geographic coordinate file and the second corrected geographic coordinate file, and removing repeated point pairs to serve as a third corrected geographic coordinate file for correcting the target satellite image.
The correction module 23 is configured to perform correction accuracy evaluation on an image corrected by using a preset scheme on a target satellite image, so as to screen out an optimal correction scheme, and correct the target satellite image according to the optimal correction scheme.
Further, the correction module 23 is, in particular,
respectively using three corrected geographic coordinate files obtained by the three schemes for geometric correction of the target image by using a quadratic polynomial method to obtain target satellite images corrected by different schemes;
extracting the homonymous points of the three corrected target satellite images and the corresponding MODIS images by using a scale invariant feature transform algorithm, calculating Euclidean distances between the point pairs, evaluating correction precision after geometric correction of different schemes, and screening out an optimal correction scheme;
and performing geometric correction on the target satellite image by using the optimal correction scheme.
Firstly, obtaining MODIS data of a target satellite on the same running day in batches, and performing data preprocessing on the MODIS data to obtain preprocessed MODIS data; sequentially carrying out primary screening and secondary screening on the preprocessed MODIS data to extract homonymous point pairs of the target satellite and the MODIS data; extracting the coordinates of the homonymous points of the homonymous point pairs by an integer point image and image fractional enhancement method to obtain a corrected geographic coordinate file; and carrying out correction precision evaluation on the corrected geographic coordinate file so as to screen out an optimal correction scheme, and correcting the satellite image of the target satellite according to the optimal correction scheme. Compared with the prior art, the method can reduce low positioning precision of polar region minisatellite, realize automatic preferential screening of the registration reference image of the heterogeneous sensor, increase the selection of the homonymy point based on the characteristics and meet the actual application requirement.
An embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program; wherein the computer program controls, when running, the apparatus on which the computer readable storage medium is located to execute the above satellite image correction method.
An embodiment of the present invention further provides a terminal device, as shown in fig. 6, which is a block diagram of a preferred embodiment of the terminal device provided in the present invention, the terminal device includes a processor 10, a memory 20, and a computer program stored in the memory 20 and configured to be executed by the processor 10, and the processor 10 implements the above-mentioned satellite image correction method when executing the computer program.
Preferably, the computer program can be divided into one or more modules/units (e.g. computer program 1, computer program 2,) which are stored in the memory 20 and executed by the processor 10 to accomplish the present invention. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, which are used for describing the execution process of the computer program in the terminal device.
The Processor 10 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf Programmable Gate Array (FPGA) or other Programmable logic device, a discrete Gate or transistor logic device, a discrete hardware component, etc., the general purpose Processor may be a microprocessor, or the Processor 10 may be any conventional Processor, the Processor 10 is a control center of the terminal device, and various interfaces and lines are used to connect various parts of the terminal device.
The memory 20 mainly includes a program storage area that may store an operating system, an application program required for at least one function, and the like, and a data storage area that may store related data and the like. In addition, the memory 20 may be a high speed random access memory, may also be a non-volatile memory, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card (Flash Card), and the like, or the memory 20 may also be other volatile solid state memory devices.
It should be noted that the terminal device may include, but is not limited to, a processor and a memory, and those skilled in the art will understand that the structural block diagram of fig. 6 is only an example of the terminal device, and does not constitute a limitation to the terminal device, and may include more or less components than those shown, or combine some components, or different components.
To sum up, according to the method, the system, the storage medium and the terminal device for automatically correcting the geometry of the satellite image provided by the embodiment of the invention, firstly, MODIS data of a target satellite on the same day of operation are obtained in batches, and the MODIS data are subjected to data preprocessing to obtain preprocessed MODIS data; sequentially carrying out primary screening and secondary screening on the preprocessed MODIS data to extract homonymous point pairs of the target satellite and the MODIS data; extracting the coordinates of the homonymous points of the homonymous point pairs by an integer point image and image fractional enhancement method to obtain a corrected geographic coordinate file; and carrying out correction precision evaluation on the corrected geographic coordinate file so as to screen out an optimal correction scheme, and correcting the satellite image of the target satellite according to the optimal correction scheme. Compared with the prior art, the method can reduce low positioning precision of polar region minisatellite, realize automatic preferential screening of the registration reference image of the heterogeneous sensor, increase the selection of the homonymy point based on the characteristics and meet the actual application requirement.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.
Claims (10)
1. An automatic geometric correction method for satellite images, which is applied to a minisatellite in a polar region, and is characterized by comprising the following steps:
preferentially screening a reference image for the automatic geometric correction of the target satellite image from three aspects of image acquisition time distance, space coverage degree and number of homonymous point pairs to obtain an optimal reference image for the geometric correction of the target satellite image;
extracting homonymous points of the target satellite image and the MODIS image by adopting a preset scheme to obtain a geographic coordinate file for geometric correction of each scheme; the preset scheme comprises the steps of extracting homonymous points from a whole scene image, dividing the image into four parts and nine parts, carrying out image enhancement processing, and then extracting homonymous points;
and carrying out correction precision evaluation on the image corrected by the target satellite image by adopting a preset scheme so as to screen out an optimal correction scheme, and correcting the target satellite image according to the optimal correction scheme.
2. The method for automatically correcting the geometric shape of a satellite image according to claim 1, wherein the MODIS image for the automatic geometric correction of the target satellite image is preferentially selected from three aspects of image acquisition time distance, spatial coverage degree and number of homonymous points, so as to obtain the optimal reference image for the geometric correction of the target satellite image, specifically:
obtaining MODIS images of a target satellite image which is cloudy in the day in batches, and creating a first data index table of the target satellite image and the MODIS images;
preprocessing a target satellite image and an MODIS image, and obtaining the MODIS image which is consistent with the target satellite image in geographic range by adopting consistent south pole projection and 250m resolution;
preliminarily screening the MODIS images, screening images with the effective data range covering more than 80% of the target images, and creating a second data index table of the target satellite images and the MODIS images;
and carrying out secondary screening on the MODIS images, extracting homonymous points of a target satellite and an MODIS image pair according to the second data index table by a feature matching method, sorting and preferentially selecting corresponding MODIS images according to the number of homonymous point pairs, and if the multi-scene MODIS data still have the same number of homonymous point pairs, taking the MODIS image close to the shooting time of the target satellite image as a final registration reference image and obtaining a third data index table for geometric correction.
3. The method for automatically correcting the geometric shape of the satellite images according to claim 2, wherein the preset scheme is adopted to extract the homonymous points of the target satellite image and the MODIS image so as to obtain the geographic coordinate files of each scheme for geometric correction, and specifically, the method comprises the following steps:
the first scheme is as follows: extracting homonymous point pairs from a target satellite panoramic image and a corresponding MODIS panoramic image through a scale invariant feature transform algorithm, sorting according to Euclidean distances of the homonymous point pairs, removing 10% of points with the largest Euclidean distances, and outputting real geographic coordinate files of the rest homonymous point pairs as first corrected geographic coordinate files for correcting the target satellite image;
scheme II: dividing a target image into four parts according to rules, cutting the MODIS image corresponding to 40 pixels extending from each part, performing piecewise linear stretching enhancement processing on each image pair, extracting dotted pairs through a scale invariant feature transform algorithm, removing points with the maximum Euclidean distance of 30%, outputting real geographic coordinate files of the rest dotted points, combining the coordinate file and a first corrected geographic coordinate file, and removing repeated point pairs to serve as a second corrected geographic coordinate file for correcting the target satellite image;
the third scheme is as follows: and dividing the target image into nine parts according to rules, obtaining real geographic coordinate files of the other same-name points after the points with the maximum Euclidean distance of 30% are removed by the same method as the scheme II, combining the coordinate files with the first corrected geographic coordinate file and the second corrected geographic coordinate file, and removing repeated point pairs to serve as a third corrected geographic coordinate file for correcting the target satellite image.
4. The method for automatically geometrically correcting satellite images according to claim 3, wherein the images of the target satellite images corrected by different methods are subjected to correction accuracy evaluation to screen out an optimal correction scheme, and the target satellite images are corrected according to the optimal correction scheme, specifically:
respectively using three corrected geographic coordinate files obtained by the three schemes for geometric correction of the target image by using a quadratic polynomial method to obtain target satellite images corrected by different schemes;
extracting the homonymous points of the three corrected target satellite images and the corresponding MODIS images by using a scale invariant feature transform algorithm, calculating Euclidean distances between the point pairs, evaluating correction precision after geometric correction of different schemes, and screening out an optimal correction scheme;
and performing geometric correction on the target satellite image by using the optimal correction scheme.
5. An automated geometric correction system for satellite imagery, the system comprising:
the screening module is used for preferentially screening the reference image for the automatic geometric correction of the target satellite image from three aspects of image acquisition time, space coverage degree and number of homonymous point pairs so as to obtain the optimal reference image for the geometric correction of the target satellite image;
the extraction module is used for extracting the homonymous points of the target satellite image and the MODIS image by adopting a preset scheme so as to obtain a geographic coordinate file of each scheme for geometric correction; the preset scheme comprises the steps of extracting homonymous points from a whole scene image, dividing the image into four parts and nine parts, carrying out image enhancement processing, and then extracting homonymous points;
and the correction module is used for carrying out correction precision evaluation on the image corrected by the target satellite image by adopting a preset scheme so as to screen out an optimal correction scheme and correcting the target satellite image according to the optimal correction scheme.
6. The satellite imagery automated geometry correction system of claim 5, wherein the filtering module, particularly configured to,
obtaining MODIS images of a target satellite image which is cloudy in the day in batches, and creating a first data index table of the target satellite image and the MODIS images;
preprocessing a target satellite image and an MODIS image, and obtaining the MODIS image which is consistent with the target satellite image in geographic range by adopting consistent south pole projection and 250m resolution;
preliminarily screening the MODIS images, screening images with the effective data range covering more than 80% of the target images, and creating a second data index table of the target satellite images and the MODIS images;
and carrying out secondary screening on the MODIS images, extracting homonymous points of a target satellite and an MODIS image pair according to the second data index table by a feature matching method, sorting and preferentially selecting corresponding MODIS images according to the number of homonymous point pairs, and if the multi-scene MODIS data still have the same number of homonymous point pairs, taking the MODIS image close to the shooting time of the target satellite image as a final registration reference image and obtaining a third data index table for geometric correction.
7. The satellite imagery automated geometry correction system of claim 6, wherein the extraction module, particularly to,
the first scheme is as follows: extracting homonymous point pairs from a target satellite panoramic image and a corresponding MODIS panoramic image through a scale invariant feature transform algorithm, sorting according to Euclidean distances of the homonymous point pairs, removing 10% of points with the largest Euclidean distances, and outputting real geographic coordinate files of the rest homonymous point pairs as first corrected geographic coordinate files for correcting the target satellite image;
scheme II: dividing a target image into four parts according to rules, cutting the MODIS image corresponding to 40 pixels extending from each part, performing piecewise linear stretching enhancement processing on each image pair, extracting dotted pairs through a scale invariant feature transform algorithm, removing points with the maximum Euclidean distance of 30%, outputting real geographic coordinate files of the rest dotted points, combining the coordinate file and a first corrected geographic coordinate file, and removing repeated point pairs to serve as a second corrected geographic coordinate file for correcting the target satellite image;
the third scheme is as follows: and dividing the target image into nine parts according to rules, obtaining real geographic coordinate files of the other same-name points after the points with the maximum Euclidean distance of 30% are removed by the same method as the scheme II, combining the coordinate files with the first corrected geographic coordinate file and the second corrected geographic coordinate file, and removing repeated point pairs to serve as a third corrected geographic coordinate file for correcting the target satellite image.
8. The satellite imagery automated geometry correction system of claim 7, wherein the correction module, particularly configured to,
respectively using three corrected geographic coordinate files obtained by the three schemes for geometric correction of the target image by using a quadratic polynomial method to obtain target satellite images corrected by different schemes;
extracting the homonymous points of the three corrected target satellite images and the corresponding MODIS images by using a scale invariant feature transform algorithm, calculating Euclidean distances between the point pairs, evaluating correction precision after geometric correction of different schemes, and screening out an optimal correction scheme;
and performing geometric correction on the target satellite image by using the optimal correction scheme.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored computer program; wherein the computer program controls the device on which the computer readable storage medium is located to execute the satellite imagery automated geometric correction method of any one of claims 1 to 4 when executed.
10. A terminal device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the satellite imagery automated geometric correction method of any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110266590.2A CN113160071B (en) | 2021-03-11 | 2021-03-11 | Satellite image automatic geometric correction method, system, medium and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110266590.2A CN113160071B (en) | 2021-03-11 | 2021-03-11 | Satellite image automatic geometric correction method, system, medium and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113160071A true CN113160071A (en) | 2021-07-23 |
| CN113160071B CN113160071B (en) | 2023-11-07 |
Family
ID=76886800
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110266590.2A Active CN113160071B (en) | 2021-03-11 | 2021-03-11 | Satellite image automatic geometric correction method, system, medium and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113160071B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663028A (en) * | 2012-03-23 | 2012-09-12 | 北京师范大学 | Method suitable for fast spatially-indexing global digital elevation model and remote sensing image data |
| CN103034988A (en) * | 2012-12-18 | 2013-04-10 | 武汉大学 | Space-time quantitative remote sensing fusion method of arbitrary number of sensors |
| CN105046251A (en) * | 2015-08-04 | 2015-11-11 | 中国资源卫星应用中心 | Automatic ortho-rectification method based on remote-sensing image of environmental No.1 satellite |
| CN107657597A (en) * | 2017-10-19 | 2018-02-02 | 中国科学院遥感与数字地球研究所 | Cross-platform moon base earth observation image automatic geometric correction method |
| US20180174312A1 (en) * | 2016-12-21 | 2018-06-21 | The Boeing Company | Method and apparatus for raw sensor image enhancement through georegistration |
| CN109903352A (en) * | 2018-12-24 | 2019-06-18 | 中国科学院遥感与数字地球研究所 | A kind of seamless orthography production method in the big region of satellite remote-sensing image |
| WO2019157348A1 (en) * | 2018-02-09 | 2019-08-15 | The Board Of Trustees Of The University Of Illinois | A system and method to fuse multiple sources of optical data to generate a high-resolution, frequent and cloud-/gap-free surface reflectance product |
| US20190251678A1 (en) * | 2017-09-30 | 2019-08-15 | Institute Of Remote Sensing And Digital Earth, Chinese Academy Of Sciences | Automatic cross-platform geometric correction method for moon-based earth observation image |
| CN111354054A (en) * | 2020-03-13 | 2020-06-30 | 中山大学 | Polar region visible light remote sensing self-adaptive mapping method |
| CN111598772A (en) * | 2020-04-03 | 2020-08-28 | 国家卫星气象中心(国家空间天气监测预警中心) | Polar orbit meteorological satellite image display system and display method |
| CN112381864A (en) * | 2020-12-08 | 2021-02-19 | 兰州交通大学 | Multi-source multi-scale high-resolution remote sensing image automatic registration technology based on antipodal geometry |
| CN112419350A (en) * | 2020-11-20 | 2021-02-26 | 武汉大学 | Remote sensing image automatic geometric registration method and system based on ground object boundary information |
-
2021
- 2021-03-11 CN CN202110266590.2A patent/CN113160071B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102663028A (en) * | 2012-03-23 | 2012-09-12 | 北京师范大学 | Method suitable for fast spatially-indexing global digital elevation model and remote sensing image data |
| CN103034988A (en) * | 2012-12-18 | 2013-04-10 | 武汉大学 | Space-time quantitative remote sensing fusion method of arbitrary number of sensors |
| CN105046251A (en) * | 2015-08-04 | 2015-11-11 | 中国资源卫星应用中心 | Automatic ortho-rectification method based on remote-sensing image of environmental No.1 satellite |
| US20180174312A1 (en) * | 2016-12-21 | 2018-06-21 | The Boeing Company | Method and apparatus for raw sensor image enhancement through georegistration |
| US20190251678A1 (en) * | 2017-09-30 | 2019-08-15 | Institute Of Remote Sensing And Digital Earth, Chinese Academy Of Sciences | Automatic cross-platform geometric correction method for moon-based earth observation image |
| CN107657597A (en) * | 2017-10-19 | 2018-02-02 | 中国科学院遥感与数字地球研究所 | Cross-platform moon base earth observation image automatic geometric correction method |
| WO2019157348A1 (en) * | 2018-02-09 | 2019-08-15 | The Board Of Trustees Of The University Of Illinois | A system and method to fuse multiple sources of optical data to generate a high-resolution, frequent and cloud-/gap-free surface reflectance product |
| CN109903352A (en) * | 2018-12-24 | 2019-06-18 | 中国科学院遥感与数字地球研究所 | A kind of seamless orthography production method in the big region of satellite remote-sensing image |
| CN111354054A (en) * | 2020-03-13 | 2020-06-30 | 中山大学 | Polar region visible light remote sensing self-adaptive mapping method |
| CN111598772A (en) * | 2020-04-03 | 2020-08-28 | 国家卫星气象中心(国家空间天气监测预警中心) | Polar orbit meteorological satellite image display system and display method |
| CN112419350A (en) * | 2020-11-20 | 2021-02-26 | 武汉大学 | Remote sensing image automatic geometric registration method and system based on ground object boundary information |
| CN112381864A (en) * | 2020-12-08 | 2021-02-19 | 兰州交通大学 | Multi-source multi-scale high-resolution remote sensing image automatic registration technology based on antipodal geometry |
Non-Patent Citations (3)
| Title |
|---|
| Y. ZHANG 等: "Automated radiation and geometric correction processing method of polar-observing cubesat BNU-1 images", 《AGU FALL MEETING 2020 ABSTRACTS》, pages 088 - 0004 * |
| YING ZHANG 等: "Accuracy evaluation on geolocation of the Chinese first polar microsatellite (Ice pathfinder) imagery", 《REMOTE SENSING》, vol. 13, no. 4278, pages 1 - 20 * |
| 张卓宇 等: ""京师一号"极地小卫星宽幅影像辐射校正方法研究", 《北京师范大学学报(自然科学版)》, vol. 59, no. 01, pages 104 - 112 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113160071B (en) | 2023-11-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111126125B (en) | Method, device, equipment and readable storage medium for extracting target text in certificate | |
| CN109543489B (en) | Positioning method and device based on two-dimensional code and storage medium | |
| CN109784250B (en) | Positioning method and device of automatic guide trolley | |
| KR20130019430A (en) | Daisy descriptor generation from precomputed scale - space | |
| CN108961180B (en) | Infrared image enhancement method and system | |
| US20110293175A1 (en) | Image processing apparatus and method | |
| CN111028170A (en) | Image processing method, image processing apparatus, electronic device, and readable storage medium | |
| CN111369605A (en) | Infrared and visible light image registration method and system based on edge features | |
| CN109754423B (en) | Method and equipment for extracting coverage area of leaf scab | |
| CN111275616B (en) | Low-altitude aerial image splicing method and device | |
| CN113963009B (en) | Local self-attention image processing method and system based on deformable block | |
| CN114092326A (en) | Web map tile updating processing method and device | |
| CN111161348A (en) | Monocular camera-based object pose estimation method, device and equipment | |
| CN111062341B (en) | Video image area classification method, device, equipment and storage medium | |
| Senturk et al. | Seam carving based image retargeting: A survey | |
| CN113160071A (en) | Satellite image automatic geometric correction method, system, medium and terminal equipment | |
| CN110245674B (en) | Template matching method, device, equipment and computer storage medium | |
| CN109785367B (en) | Method and device for filtering foreign points in three-dimensional model tracking | |
| CN114092695B (en) | ROI extraction method and device based on segmentation model | |
| CN116109535A (en) | Image fusion method, device and computer readable storage medium | |
| CN115690139A (en) | Tracking processing method and device for image edge and server | |
| CN110689586A (en) | Tongue image identification method in traditional Chinese medicine intelligent tongue diagnosis and portable correction color card used for same | |
| CN112419407B (en) | Cloud cluster displacement vector calculation method and device based on cloud cluster edge recognition | |
| CN114913105A (en) | Laser point cloud fusion method and device, server and computer readable storage medium | |
| CN116170569A (en) | Hyperspectral image forward shooting correction method and device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |