CN109993104A - A kind of change detecting method of remote sensing images object hierarchy - Google Patents
A kind of change detecting method of remote sensing images object hierarchy Download PDFInfo
- Publication number
- CN109993104A CN109993104A CN201910246785.3A CN201910246785A CN109993104A CN 109993104 A CN109993104 A CN 109993104A CN 201910246785 A CN201910246785 A CN 201910246785A CN 109993104 A CN109993104 A CN 109993104A
- Authority
- CN
- China
- Prior art keywords
- pixel
- image
- formula
- remote sensing
- value
- 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 17
- 230000008859 change Effects 0.000 title claims abstract description 12
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 12
- 238000010191 image analysis Methods 0.000 claims abstract description 7
- 238000003703 image analysis method Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims 1
- 238000006243 chemical reaction Methods 0.000 abstract description 3
- 230000003044 adaptive effect Effects 0.000 abstract description 2
- 238000001514 detection method Methods 0.000 abstract description 2
- 238000005516 engineering process Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 5
- VMXUWOKSQNHOCA-UKTHLTGXSA-N ranitidine Chemical compound [O-][N+](=O)\C=C(/NC)NCCSCC1=CC=C(CN(C)C)O1 VMXUWOKSQNHOCA-UKTHLTGXSA-N 0.000 description 5
- 230000018109 developmental process Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000003595 spectral effect Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 241000908464 Geobia Species 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 230000001149 cognitive effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 210000000697 sensory organ Anatomy 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000013456 study Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of change detecting methods of remote sensing images object hierarchy, specifically include image preliminary treatment, mean filter obtains calculating and the image analysis based on convolutional neural networks;The invention proposes an adaptive k value calculating method, the invention proposes a kind of change detecting methods of remote sensing images object hierarchy, it is whole that preliminary figure is obtained based on the full convolutional neural networks model of depth, and image is finally carried out by binary conversion treatment based on corresponding algorithm, to realize the digitized processing of image, the variation for carrying out the later period to image that the digitized processing based on image can be convenient at this time detects, to improve the precision and accuracy of detection.
Description
Technical field
The present invention relates to field of image processing, more particularly, to a kind of change detecting method of remote sensing images object hierarchy,
It is mainly used for gray processing and binary conversion treatment to traditional remote sensing images.
Background technique
The appearance of remote sensing technology changes the mode that the mankind observe the earth;The acquisition of the remote sensing data of repeatability, so that
The mankind are provided with the new way for periodically understanding wide area ground surface environment dynamic change.With the development of remote sensing technology, optical remote sensing
The spatial resolution of image is increased to meter level from hundred initial meter levels, and even higher, revisiting period has also greatly shortened.With
The development of image technique, to be obtained using the spatial information in remote sensing images as the image analysis method towards geographic object of target
Rapid development is arrived.
At this stage, meter level/sub-meter grade definition satellite remotely-sensed data is numerous, external commercial satellite data such as SPOT,
IKONOS, QuickBird, WorldView, GeoEye etc., China's high-definition remote sensing data have high score No.1, high score No. two numbers
Deng.Such satellite generally uses sun-synchronous orbit to obtain data, and satellite gravity anomaly is flexible, can be by tilting and swinging partially
From substar earth observation.Compared with Landsat satellite of early stage etc. obtains the image-forming mechanism of substar ground mulching type, rice
The spatial information that the geometry of target object is contained in grade remote sensing images is the important information source for identifying earth object.Tradition
Atural object in earth's surface scene is often simply interpreted as that there is similar lambert's system spectral reflectance to emit by Remote Sensing Information Extraction method
Structure, shortage further interpret process to scene space information.However, in high-resolution remote sensing image, the sky of atural object
Between structure externalize the set of multiple adjacent pixels, traditional Remote Sensing Information Extraction means are faced with new challenges.
The remote sensing platforms such as high-resolution Commercial Remote Sensing Satellites abundant, a plurality of types of unmanned vehicles, manned aviation and
Its sensor respectively has advantage, and the acquisition capability of high-definition remote sensing data obtains considerable raising.Remote sensing spatial resolution mentions
Height, to we provide a kind of sense organ cognitions on the spot in person, and current Remote Sensing Image Processing Technology is a lack of mankind's view
Feel the vision processing mechanism in perception.Current remote sensing image processing method is mainly based upon pixel spectral signature, multiple
Our requirements to technology itself are but not achieved in geoscience applications in miscellaneous region.Remote Sensing Information Extraction ability institute visual not as good as ordinary person
The abundant information of acquisition.Therefrom low resolution is increased to sub-meter grade high-resolution, the image feature hair of ground object target to remote sensing image
Significant changes are given birth to, information is more abundant, is difficult to describe numerous complicated on high resolution image comprehensively using single spectral signature
Ground object target.However, the Data Analysis Services ability of complex environment remote sensing is not correspondingly improved but, it is ever-increasing
Most of data in mass remote sensing data do not obtain effective analysis and utilization.It can be seen that realizing remotely-sensed data to sky
Between information intelligent convert image Cognitive Mode be in urgent need with correlative studys such as image understandings.How high-resolution is made full use of
The priori knowledges such as the object spectrum, space structure, the image-forming information that are contained in remote sensing information establish computable remote sensing information solution
Analysis mechanism analyzes complex scene environment, is the difficult point that high-resolution remote sensing image understands.
The raising of spatial resolution, so that the atural object that we recognize is rendered as pixel set in remote sensing image.In high score
In the remote sensing image of resolution, need to go to obtain by way of image procossing the pixel set and its space that it is reflected on image
Structure.In last decade, to obtain and using pixel set in image as the various partitioning algorithms of target and object-oriented
Image analysis method (OBIA/ GEOBIA) rapidly develops, and has been a much progress of technology development for Pixel-level.With
This improves Remote Sensing Information Extraction ability using atural object spatial information and is more paid attention to simultaneously, is existed using the key of spatial information
In obtain image object, image object refer on remote sensing image with the consistent pixel set of atural object intrinsic time theory.Therefore, synthesis is distant
The all information such as spectrum and spatial information that sense image can reflect develop the remote sensing image information processing skill of image object hierarchy
Art has been fundamentally to improve the certainty of Remote Sensing Information Extraction ability, and further realize the base of remote sensing image interpretation reasoning
Plinth.
Summary of the invention
The present invention is to overcome above situation insufficient, it is desirable to provide a kind of technical solution that can solve the above problem.
A kind of change detecting method of remote sensing images object hierarchy specifically includes image preliminary treatment, mean filter obtains
Calculating and image analysis;
The initial pictures input full convolutional neural networks model of depth trained in advance is obtained the depth by S1, image preliminary treatment
Spend the probability that each pixel in the initial pictures of full convolutional neural networks model output is character pixels point, wherein institute
Stating the full convolutional neural networks model of depth is to advance with the training image of the real estate for being labeled with character to be trained to obtain;
Then the pixel in the initial pictures is classified, wherein the pixel that probability is greater than predetermined probabilities threshold value is classified as character
Pixel;Multiple trained figure layers are obtained by multiple image procossing, and training map overlay is obtained into preliminary figure;
S2, the preliminary figure for exporting step S1 carry out mean filter and obtain calculating: using Sauvola algorithm to gray level image
Threshold segmentation is carried out, as the formula:
In formula (1), D (x, y) represents the contrast value of pixel, is calculated based on formula (1);
Wherein T indicates the mean filter acquisition of current pixel, and m and s respectively indicate the gray scale of pixel within the scope of current neighborhood
Mean value and standard deviation, k are an adjustment factors, and the response for control algolithm to picture contrast, R is to gray standard deviation
An adjusting it is related with the gray scale order of image;
E (x, y) represents whether pixel is marginal point, and if it is marginal point, then the value is 1, and otherwise the value is 0, Ne representative
The number of marginal point in current pixel point contiguous range, the k value being calculated by formula (1), actually current picture
Why the contrast mean value at edge in vegetarian refreshments neighborhood does not use the contrast information of contiguous range all the points, is because of non-side
The variation of edge point local gray-value is smaller, and contrast cannot reflect the contrast distribution feature of neighborhood, bring formula (1) into formula (2)
It obtains new mean filter and obtains calculation formula:
In formula (3), when calculating threshold value other than using neighborhood of pixel points grayscale information, the comparison of contiguous range is also introduced
Information is spent, can adapt to the inconsistent situation of picture contrast;
S3, image analysis: image analysis method is as the formula (4):
G (x, y) represents the gray value of current pixel point, and T is that the mean filter being calculated using formula (3) is obtained, and Ne is represented
The number of marginal point in current pixel point contiguous range, only when the gray value of the pixel less than its mean filter obtain, and
When the number of the neighborhood of pixel points range inward flange pixel is greater than neighborhood diameter (current point is in adjacent edges), the pixel is determined
Point is foreground pixel, and otherwise the pixel is background pixel, traverses all the points on image, calculates mean filter and obtains and count
The number of marginal point in contiguous range completes the binaryzation of image using formula (4).
Beneficial effects of the present invention: the invention proposes an adaptive k value calculating method, the invention proposes one
The change detecting method of kind remote sensing images object hierarchy, it is whole that preliminary figure is obtained based on the full convolutional neural networks model of depth
Shape, and image is finally carried out by binary conversion treatment based on corresponding algorithm, to realize the digitized processing of image, it is based at this time
The variation for carrying out the later period to image that the digitized processing of image can be convenient detects, to improve the precision of detection and accurate
Property.
Additional aspect and advantage of the invention will be set forth in part in the description, and will partially become from the following description
Obviously, or practice through the invention is recognized.
Detailed description of the invention
Fig. 1 is inventive algorithm process.
Specific embodiment
In the embodiment of the present invention, a kind of change detecting method of remote sensing images object hierarchy specifically includes image and tentatively locates
Reason, mean filter obtain calculating and image analysis;
The initial pictures input full convolutional neural networks model of depth trained in advance is obtained the depth by S1, image preliminary treatment
Spend the probability that each pixel in the initial pictures of full convolutional neural networks model output is character pixels point, wherein institute
Stating the full convolutional neural networks model of depth is to advance with the training image of the real estate for being labeled with character to be trained to obtain;
Then the pixel in the initial pictures is classified, wherein the pixel that probability is greater than predetermined probabilities threshold value is classified as character
Pixel;Multiple trained figure layers are obtained by multiple image procossing, and training map overlay is obtained into preliminary figure;
S2, the preliminary figure for exporting step S1 carry out mean filter and obtain calculating: using Sauvola algorithm to gray level image
Threshold segmentation is carried out, as the formula:
In formula (1), D (x, y) represents the contrast value of pixel, is calculated based on formula (1);
Wherein T indicates the mean filter acquisition of current pixel, and m and s respectively indicate the gray scale of pixel within the scope of current neighborhood
Mean value and standard deviation, k are an adjustment factors, and the response for control algolithm to picture contrast, R is to gray standard deviation
An adjusting it is related with the gray scale order of image;
E (x, y) represents whether pixel is marginal point, and if it is marginal point, then the value is 1, and otherwise the value is 0, Ne representative
The number of marginal point in current pixel point contiguous range, the k value being calculated by formula (1), actually current picture
Why the contrast mean value at edge in vegetarian refreshments neighborhood does not use the contrast information of contiguous range all the points, is because of non-side
The variation of edge point local gray-value is smaller, and contrast cannot reflect the contrast distribution feature of neighborhood, bring formula (1) into formula (2)
It obtains new mean filter and obtains calculation formula:
In formula (3), when calculating threshold value other than using neighborhood of pixel points grayscale information, the comparison of contiguous range is also introduced
Information is spent, can adapt to the inconsistent situation of picture contrast;
S3, image analysis: image analysis method is as the formula (4):
G (x, y) represents the gray value of current pixel point, and T is that the mean filter being calculated using formula (3) is obtained, and Ne is represented
The number of marginal point in current pixel point contiguous range, only when the gray value of the pixel less than its mean filter obtain, and
When the number of the neighborhood of pixel points range inward flange pixel is greater than neighborhood diameter (current point is in adjacent edges), the pixel is determined
Point is foreground pixel, and otherwise the pixel is background pixel, traverses all the points on image, calculates mean filter and obtains and count
The number of marginal point in contiguous range completes the binaryzation of image using formula (4).
It is obvious to a person skilled in the art that invention is not limited to the details of the above exemplary embodiments, Er Qie
In the case where without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Benefit requires rather than above description limits, it is intended that all by what is fallen within the meaning and scope of the equivalent elements of the claims
Variation is included within the present invention.
Claims (3)
1. a kind of change detecting method of remote sensing images object hierarchy, which is characterized in that specifically include image preliminary treatment, mean value
Filtering obtains calculating and image analysis;
The initial pictures input full convolutional neural networks model of depth trained in advance is obtained the depth by S1, image preliminary treatment
Spend the probability that each pixel in the initial pictures of full convolutional neural networks model output is character pixels point, wherein institute
Stating the full convolutional neural networks model of depth is to advance with the training image of the real estate for being labeled with character to be trained to obtain;
Then the pixel in the initial pictures is classified, wherein the pixel that probability is greater than predetermined probabilities threshold value is classified as character
Pixel;Multiple trained figure layers are obtained by multiple image procossing, and training map overlay is obtained into preliminary figure;
S2, the preliminary figure for exporting step S1 carry out mean filter and obtain calculating: using Sauvola algorithm to gray level image
Threshold segmentation is carried out, as the formula:
In formula (1), D (x, y) represents the contrast value of pixel, is calculated based on formula (1);
Wherein T indicates the mean filter acquisition of current pixel, and m and s respectively indicate the gray scale of pixel within the scope of current neighborhood
Mean value and standard deviation, Ne represent the number of the marginal point in current pixel point contiguous range, the k being calculated by formula (1)
Value brings formula (1) into formula (2) and obtains new mean filter acquisition calculation formula:
In formula (3), when calculating threshold value other than using neighborhood of pixel points grayscale information, the comparison of contiguous range is also introduced
Information is spent, can adapt to the inconsistent situation of picture contrast;
S3, image analysis: image analysis method is as the formula (4):
;
G (x, y) represents the gray value of current pixel point, and T is that the mean filter being calculated using formula (3) is obtained, and determines the picture
Vegetarian refreshments is foreground pixel, and otherwise the pixel is background pixel, traverses all the points on image, calculates mean filter and obtains and unite
The number for counting marginal point in contiguous range completes the binaryzation of image using formula (4).
2. the change detecting method of remote sensing images object hierarchy according to claim 1, which is characterized in that k is a tune
Coefficient, the response for control algolithm to picture contrast are saved, R is the gray scale of an adjusting and image to gray standard deviation
Order is related;E (x, y) represents whether pixel is marginal point, and if it is marginal point, then the value is 1, and otherwise the value is 0.
3. the change detecting method of remote sensing images object hierarchy according to claim 1, Ne represent current pixel vertex neighborhood
The number of marginal point in range, only when the gray value of the pixel is obtained less than its mean filter, and the neighborhood of pixel points model
When enclosing the number of inward flange pixel greater than neighborhood diameter (i.e. current point is in adjacent edges).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910246785.3A CN109993104B (en) | 2019-03-29 | 2019-03-29 | Method for detecting change of object level of remote sensing image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910246785.3A CN109993104B (en) | 2019-03-29 | 2019-03-29 | Method for detecting change of object level of remote sensing image |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109993104A true CN109993104A (en) | 2019-07-09 |
CN109993104B CN109993104B (en) | 2022-09-16 |
Family
ID=67131800
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910246785.3A Expired - Fee Related CN109993104B (en) | 2019-03-29 | 2019-03-29 | Method for detecting change of object level of remote sensing image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109993104B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114676907A (en) * | 2022-01-17 | 2022-06-28 | 中国地质大学(北京) | Regional geological disaster early warning method and device, storage medium and equipment |
CN116012769A (en) * | 2022-09-07 | 2023-04-25 | 广东海洋大学 | Shipborne radar image oil spill identification method based on GLCM and K-means clustering algorithm |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016207875A1 (en) * | 2015-06-22 | 2016-12-29 | Photomyne Ltd. | System and method for detecting objects in an image |
CA2995850A1 (en) * | 2015-08-31 | 2017-03-09 | Ryan Kottenstette | Systems and methods for analyzing remote sensing imagery |
CN108573276A (en) * | 2018-03-12 | 2018-09-25 | 浙江大学 | A kind of change detecting method based on high-resolution remote sensing image |
WO2018214195A1 (en) * | 2017-05-25 | 2018-11-29 | 中国矿业大学 | Remote sensing imaging bridge detection method based on convolutional neural network |
CN108932303A (en) * | 2018-06-12 | 2018-12-04 | 中国电子科技集团公司第二十八研究所 | A kind of distribution visual remote sensing image Detection dynamic target and analysis system |
CN109064496A (en) * | 2018-08-15 | 2018-12-21 | 贵州工程应用技术学院 | A kind of change detecting method of remote sensing images object hierarchy |
-
2019
- 2019-03-29 CN CN201910246785.3A patent/CN109993104B/en not_active Expired - Fee Related
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2016207875A1 (en) * | 2015-06-22 | 2016-12-29 | Photomyne Ltd. | System and method for detecting objects in an image |
CA2995850A1 (en) * | 2015-08-31 | 2017-03-09 | Ryan Kottenstette | Systems and methods for analyzing remote sensing imagery |
WO2018214195A1 (en) * | 2017-05-25 | 2018-11-29 | 中国矿业大学 | Remote sensing imaging bridge detection method based on convolutional neural network |
CN108573276A (en) * | 2018-03-12 | 2018-09-25 | 浙江大学 | A kind of change detecting method based on high-resolution remote sensing image |
CN108932303A (en) * | 2018-06-12 | 2018-12-04 | 中国电子科技集团公司第二十八研究所 | A kind of distribution visual remote sensing image Detection dynamic target and analysis system |
CN109064496A (en) * | 2018-08-15 | 2018-12-21 | 贵州工程应用技术学院 | A kind of change detecting method of remote sensing images object hierarchy |
Non-Patent Citations (3)
Title |
---|
曲景影等: "基于CNN模型的高分辨率遥感图像目标识别", 《国外电子测量技术》 * |
欧攀等: "基于卷积神经网络的遥感图像目标检测", 《激光与光电子学进展》 * |
罗仙仙等: "深度学习方法用于遥感图像处理的研究进展", 《泉州师范学院学报》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114676907A (en) * | 2022-01-17 | 2022-06-28 | 中国地质大学(北京) | Regional geological disaster early warning method and device, storage medium and equipment |
CN114676907B (en) * | 2022-01-17 | 2022-09-20 | 中国地质大学(北京) | Regional geological disaster early warning method and device, storage medium and equipment |
CN116012769A (en) * | 2022-09-07 | 2023-04-25 | 广东海洋大学 | Shipborne radar image oil spill identification method based on GLCM and K-means clustering algorithm |
Also Published As
Publication number | Publication date |
---|---|
CN109993104B (en) | 2022-09-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111798475B (en) | Indoor environment 3D semantic map construction method based on point cloud deep learning | |
CN111383192A (en) | SAR-fused visible light remote sensing image defogging method | |
CN111814771B (en) | Image processing method and device | |
CN111815665B (en) | Single image crowd counting method based on depth information and scale perception information | |
CN114724120B (en) | Vehicle target detection method and system based on radar vision semantic segmentation adaptive fusion | |
CN108428220A (en) | Satellite sequence remote sensing image sea island reef region automatic geometric correction method | |
CN109034233A (en) | A kind of high-resolution remote sensing image multi classifier combination classification method of combination OpenStreetMap | |
CN102446278B (en) | Multitemporal remote sensing image-based subpixel positioning method | |
CN113239830A (en) | Remote sensing image cloud detection method based on full-scale feature fusion | |
US20230281913A1 (en) | Radiance Fields for Three-Dimensional Reconstruction and Novel View Synthesis in Large-Scale Environments | |
CN116310883B (en) | Agricultural disaster prediction method based on remote sensing image space-time fusion and related equipment | |
CN109993104A (en) | A kind of change detecting method of remote sensing images object hierarchy | |
CN116311412A (en) | Mask wearing detection method integrating 3D attention mechanism and cavity convolution | |
CN115471749A (en) | Multi-view multi-scale target identification method and system for extraterrestrial detection unsupervised learning | |
CN114266947A (en) | Classification method and device based on fusion of laser point cloud and visible light image | |
CN112883823A (en) | Land cover category sub-pixel positioning method based on multi-source remote sensing data fusion | |
CN116129234A (en) | Attention-based 4D millimeter wave radar and vision fusion method | |
Han et al. | Hypothetical visible bands of advanced meteorological imager onboard the geostationary Korea multi-purpose satellite-2A using data-to-data translation | |
CN110598702A (en) | Preparation method, system, equipment and medium of ship target detection data set | |
CN116310832A (en) | Remote sensing image processing method, device, equipment, medium and product | |
CN115984712A (en) | Multi-scale feature-based remote sensing image small target detection method and system | |
CN113624691B (en) | Spectral image super-resolution mapping method based on space-spectrum correlation | |
CN114155425A (en) | Weak and small target detection method based on Gaussian Markov random field motion direction estimation | |
CN113762271A (en) | SAR image semantic segmentation method and system based on irregular convolution kernel neural network model | |
CN109583484B (en) | Automatic selection method for three-type sea area landmark points |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20220916 |
|
CF01 | Termination of patent right due to non-payment of annual fee |