CN113627331A - High-resolution image road extraction method based on extended road shape index - Google Patents
High-resolution image road extraction method based on extended road shape index Download PDFInfo
- Publication number
- CN113627331A CN113627331A CN202110914630.XA CN202110914630A CN113627331A CN 113627331 A CN113627331 A CN 113627331A CN 202110914630 A CN202110914630 A CN 202110914630A CN 113627331 A CN113627331 A CN 113627331A
- Authority
- CN
- China
- Prior art keywords
- road
- image
- spatial
- extraction
- extended
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
Abstract
The invention provides a high-resolution image road extraction method based on an extended road shape index, which relates to the technical field of remote sensing target extraction and comprises the following modules; a preprocessing module: extracting spectral features and spatial autocorrelation features of the high-resolution remote sensing image in a mode of one-by-one image spot, extracting image spots by using a superpixel segmentation method, and calculating spectral features and spatial autocorrelation features of each image spot; the extended road shape index extraction module: after the image spots are extracted, each image spot is scanned and processed through an iterative program, and an expansion area of each image spot is obtained through an area expansion algorithm; the extraction of the spatial features of the road is expanded to multiple scales, and the problem of inaccurate description of the spatial features of the road in the traditional object-oriented road extraction method is solved by constructing a road shape index with expansibility.
Description
Technical Field
The invention relates to the technical field of remote sensing target extraction, in particular to a high-resolution image road extraction method based on an extended road shape index.
Background
Road networks are important components of traffic infrastructure and also important basic data of geographical national conditions. The integrated, timely and accurate road network has wide social demands on daily travel of people, logistics distribution of electronic commerce and national development strategy planning, plays an important role in social and economic activities, and is informationized and intelligentized as the full arrival of the internet era and the rapid development of mobile internet and artificial intelligence technology, so that the future development direction of cities is provided. Taking a smart city as an example, the method is based on a computer technology, a multimedia technology, a large-scale storage technology and an internet of things technology, takes a broadband network as a link, and applies technologies such as remote sensing, a global positioning system, a geographic information system, remote sensing, simulation-virtualization and the like to carry out multi-resolution, multi-scale, multi-space-time and multi-type three-dimensional description on the city. The digital city infrastructure is the basis of smart city construction, and as an important component part of the digital city infrastructure, the construction of a high-precision three-dimensional road network is more and more important, and due to the rapid development of the aviation and aerospace remote sensing technologies, the remote sensing technology gradually becomes an important tool for the information-based high-speed development of the current society. The remote sensing technology can rapidly and accurately acquire a large amount of ground observation data, and compared with a medium-low resolution satellite image and a high-resolution remote sensing image, the remote sensing technology has more massive and abundant ground surface Information, provides more favorable conditions for dynamic update of Geographic Information System (GIS) data and development and application of GIS, and has important significance for map update, image matching, change detection and the like. Therefore, extracting roads from high-resolution images becomes a current research focus.
At present, the surveying and mapping data production department mainly uses a manual visual interpretation method to extract roads from remote sensing images and aerial images. The method for manual visual interpretation is relatively easy to operate and has high extraction precision, and the reason is that human beings have a large amount of road prior information, including road shape characteristics, topological characteristics, contrast with other ground objects and the like; meanwhile, the artificial visual interpretation is insensitive to image noise, and can well overcome the adverse factors that buildings, shadows, vegetation on two sides of the road and the like easily influence the extraction of the road. Therefore, it is feasible to extract roads by a method of manual visual interpretation under the condition of abundant manpower and time resources; however, when a road network is required to be extracted quickly in a large area and in an emergency, the speed of manual extraction is far from meeting the requirement. At this time, a computer is needed to realize automatic or semi-automatic extraction of the road network, so that the workload of the interior work is reduced, and the working efficiency is improved. However, in practical applications, due to the complexity of roads, such as image noise, differences in road materials (cement, asphalt), buildings, vegetation and shadows, the existing road extraction algorithm has not completely solved the problem of road extraction. Therefore, extracting roads from remote sensing images is still a difficult point in the field of remote sensing, and further research is urgently needed.
Disclosure of Invention
The invention aims to provide a high-resolution image road extraction method based on an extended road shape index, and aims to solve the problems that in the prior art, the road feature use scale of the traditional road extraction method is single, and the spatial shape feature description of a road is inaccurate in the traditional object-oriented road extraction method.
In order to achieve the purpose, the invention adopts the following technical scheme: a high-resolution image road extraction method based on an extended road shape index comprises the following modules; a preprocessing module: extracting spectral features and spatial autocorrelation features of the high-resolution remote sensing image in a mode of one-by-one image spot, extracting image spots by using a superpixel segmentation method, and calculating spectral features and spatial autocorrelation features of each image spot; the extended road shape index extraction module: after image spots are extracted, each image spot is scanned and processed through an iterative program, an expansion area of each image spot is obtained through an area expansion algorithm, and space shape characteristics of an expansion road area are calculated, wherein the space shape characteristics comprise an object linear index and an average road width characteristic based on a skeleton; the road extraction module: and estimating the road probability based on a multi-scale collaborative representation method, and finally, performing road segmentation by using a graph cutting method to obtain a road surface extraction result.
In a further technical solution of the present invention, in the preprocessing module, the super-pixel segmentation method includes the steps of: initializing a seed point: uniformly distributing seed points in the image according to the set number of the super pixels; reselecting seed points in n-n neighborhoods of the seed points, calculating gradient values of all pixel points in the neighborhoods, and moving the seed points to the place with the minimum gradient in the neighborhoods, wherein n is 3; distributing a class label to each pixel point in the neighborhood around each seed point; distance measurement: the method comprises the steps of calculating the distance between each searched pixel point and the seed point respectively according to the color distance and the space distance; iterative optimization: continuously iterating the steps until the error is converged, wherein the iteration number is 10; and (3) enhancing connectivity: and newly building a marking table, wherein the elements in the table are all-1, the discontinuous superpixels and the super-pixels with undersize are redistributed to the adjacent superpixels according to the Z-shaped trend, and the traversed superpixels are distributed to the corresponding labels until all the superpixels are traversed.
The invention further adopts the technical scheme that the extended road shape index extraction module comprises two parts of region extension by taking the segmentation pattern spot as a center and spatial feature extraction based on an extended region.
The further technical scheme of the invention is that the region expansion with the segmentation pattern spot as the center comprises the following steps: introducing a first geographic law for extracting image spatial features and expressing spatial correlation by using a Moran index; the method comprises the following steps of (1) carrying out superpixel shape characteristics based on spectral characteristics, spatial characteristics and textural characteristics according to spatial extension characteristics of roads in a high-resolution remote sensing image and uniform regular characteristics of spots segmented by using a SLIC superpixel segmentation method; and the image spot region expansion algorithm comprises the following steps: inputting a segmented image spot Oc, outputting Or: a set of patches surrounding Or.
The further technical scheme of the invention is that the pattern spot region expansion algorithm specifically comprises the following steps: an initialization step: adding Oc to Or; collecting the pattern spots in contact with the topology in a container Ocon, wherein Ocon is O1, O2, O3, … … Ot; constructing a feature vector Vc based on the spectrum mean value of the pattern spots Oc, calculating a feature vector Vk of each pattern spot Ocon by the same method, wherein k is more than or equal to 1 and less than or equal to t, comparing the spectrum distance between Vc and Vk, and selecting adjacent pattern spots Os with the nearest distance from Ocon; comparing Os with Oc, if Os and Oc satisfy constraint rules R1, R2 and R3, Os is accepted as a plaque belonging to the same target source as Oc; adding Os into Or, and replacing the Oc with Os to perform the next step of extended exploration; from step 1 to step 5 is an iterative process, and when any of the three constraint rules are not satisfied, the iterative extension terminates and returns Or.
The further technical scheme of the invention is that the spatial feature extraction based on the extended area comprises the following steps: when the expansion iteration around the central patch terminates, a set of uniform and spatially continuous regions Or is output, using the skeleton-based object linearity index and the average road width for describing the spatial features of the regions Or of the patch composition.
The further technical scheme of the invention is that the road extraction module process is as follows: road extraction based on multi-scale collaborative representation and graph cut algorithm, a multi-scale collaborative representation method is selected, a test sample is represented by linear combination of training samples, the road probability of each object on a certain scale is obtained, all pixels in the same object are endowed with a likelihood value which is the same as that of the object, all road probability graphs of three different scales are fused into a whole, and finally the road possibility of each pixel is obtained; given an image I, constructing an undirected graph, and obtaining a road probability graph through a multi-scale collaborative representation method so as to segment the image.
The invention has the beneficial effects that:
aiming at the problem that the road feature using scale of the traditional road extraction method is single, the multi-scale property and the spatial structure diversity of the road in the high-resolution remote sensing image are considered, the spatial feature extraction of the road is expanded to the multi-scale, and the problem that the spatial shape feature description of the road is inaccurate in the traditional object-oriented road extraction method is solved by constructing a road shape index with expansibility.
Drawings
Fig. 1 is a flowchart of road extraction based on an extended road shape index in the present invention.
Detailed Description
The following further describes embodiments of the present invention with reference to the drawings.
As shown in fig. 1, a method for extracting a high resolution image road based on an extended road shape index includes the following modules; a preprocessing module: extracting spectral features and spatial autocorrelation features of the high-resolution remote sensing image in a mode of one-by-one image spot, extracting image spots by using a superpixel segmentation method, and calculating spectral features and spatial autocorrelation features of each image spot; the extended road shape index extraction module: after image spots are extracted, each image spot is scanned and processed through an iterative program, an expansion area of each image spot is obtained through an area expansion algorithm, and space shape characteristics of an expansion road area are calculated, wherein the space shape characteristics comprise an object linear index and an average road width characteristic based on a skeleton; the road extraction module: estimating the road probability based on a multi-scale collaborative representation method, and finally performing road segmentation by using a graph cutting method to obtain a road surface extraction result, wherein the following steps are required: image patches can be extracted using a Simple Linear Iterative Clustering algorithm-based superpixel segmentation method, which is an object-oriented segmentation method that has more available features, such as shape, size, and texture, than a single pixel, and can smooth many "salt and pepper" noises.
In the specific embodiment, aiming at the problem that the road features of the traditional road extraction method are single in use scale, the spatial feature extraction of the road is expanded to multiple scales by considering the multiscale property and the spatial structure diversity of the road in the high-resolution remote sensing image, and the problem that the spatial shape feature description of the road is inaccurate in the traditional object-oriented road extraction method is solved by constructing the road shape index with expansibility.
In another embodiment of the present invention, in the preprocessing module, the super-pixel segmentation method includes the following steps: initializing a seed point, namely a clustering center: uniformly distributing seed points in the image according to the set number of the super pixels; reselecting seed points in n-n neighborhoods of the seed points, calculating gradient values of all pixel points in the neighborhoods, and moving the seed points to the place with the minimum gradient in the neighborhoods, wherein n is 3; in order to avoid that the seed points fall on the contour boundary with larger gradient so as to avoid influencing the subsequent clustering effect, a class label is distributed to each pixel point in the neighborhood around each seed point, namely which clustering center belongs to; unlike standard k-means search through the entire graph, the search range of SLIC is limited to 2S × 2S, which can speed up algorithm convergence, distance metric: the method comprises the steps of calculating the distance between each searched pixel point and the seed point respectively according to the color distance and the space distance; iterative optimization: the above steps are iterated continuously until the error is converged, which can be understood that the clustering center of each pixel point is not changed any more, and experiments find that 10 iterations can obtain a relatively ideal effect on most pictures, so that the number of iterations is generally 10, but is not limited to 10, and can be 8, 9, 11, 12 and the like; and (3) enhancing connectivity: newly building a mark table, wherein the elements in the table are all-1, the discontinuous superpixels and the super-pixels with undersize sizes are redistributed to the adjacent superpixels according to the Z-shaped trend, namely from left to right and from top to bottom, the traversed superpixels are distributed to the corresponding labels until all the superpixels are traversed,
specifically, the extended road shape index extraction module performs region extension with the segmentation pattern as a center and extracts spatial features based on an extended region.
Specifically, the region expansion with the segmentation pattern spot as the center includes the following steps: introducing a first geographic law for extracting image spatial features and expressing spatial correlation by using a Moran index; the method comprises the following steps of (1) carrying out superpixel shape characteristics based on spectral characteristics, spatial characteristics and textural characteristics according to spatial extension characteristics of roads in a high-resolution remote sensing image and uniform regular characteristics of spots segmented by using a SLIC superpixel segmentation method; and the image spot region expansion algorithm comprises the following steps: inputting a segmented image spot Oc, outputting Or: a set of patches surrounding Or.
Specifically, the image spot region expansion algorithm specifically includes the following steps: an initialization step: adding Oc to Or; collecting the pattern spots in contact with the topology in a container Ocon, wherein Ocon is O1, O2, O3, … … Ot;
constructing a feature vector Vc based on the spectrum mean value of the pattern spots Oc, calculating a feature vector Vk of each pattern spot Ocon by the same method, wherein k is more than or equal to 1 and less than or equal to t, comparing the spectrum distance between Vc and Vk, and selecting adjacent pattern spots Os with the nearest distance from Ocon; comparing Os with Oc, if Os and Oc satisfy constraint rules R1, R2 and R3, Os is accepted as a plaque belonging to the same target source as Oc; adding Os into Or, and replacing the Oc with Os to perform the next step of extended exploration; from step 1 to step 5 is an iterative process, and when any of the three constraint rules are not satisfied, the iterative extension terminates and returns Or.
Specifically, when the expansion iteration around the central patch is terminated, a set of uniform and spatially continuous regions Or is output, and the skeleton-based object linearity index and the average road width are used to describe the spatial characteristics of the regions Or composed of the patches.
Specifically, the road extraction module comprises the following processes: road extraction based on multi-scale collaborative representation and graph cut algorithm, a multi-scale collaborative representation method is selected, a test sample is represented by linear combination of training samples, the road probability of each object on a certain scale is obtained, all pixels in the same object are endowed with a likelihood value which is the same as that of the object, all road probability graphs of three different scales are fused into a whole, and finally the road possibility of each pixel is obtained; given an image I, the GC algorithm constructs an undirected graph G ═ ν, ε, where ν represents the set of pixels in the image and ε represents the set of edges of the undirected graph between neighboring pixels. For the road extraction task, a label "1" is defined for the road category and a label "0" is defined for the non-road category. The GC attempts to minimize an objective function of the following region term and boundary term, and a road probability map is obtained by the above-mentioned multi-scale collaborative representation method, so as to segment the image.
In the description of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, e.g., as meaning a fixed connection, a removable connection, or an integral connection; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (7)
1. A high-resolution image road extraction method based on an extended road shape index is characterized by comprising the following steps: the system comprises the following modules;
a preprocessing module: extracting spectral features and spatial autocorrelation features of the high-resolution remote sensing image in a mode of one-by-one image spot, extracting image spots by using a superpixel segmentation method, and calculating spectral features and spatial autocorrelation features of each image spot;
the extended road shape index extraction module: after image spots are extracted, each image spot is scanned and processed through an iterative program, an expansion area of each image spot is obtained through an area expansion algorithm, and space shape characteristics of an expansion road area are calculated, wherein the space shape characteristics comprise an object linear index and an average road width characteristic based on a skeleton;
the road extraction module: and estimating the road probability based on a multi-scale collaborative representation method, and finally, performing road segmentation by using a graph cutting method to obtain a road surface extraction result.
2. The extended road shape index-based high resolution image road extraction method as claimed in claim 1, wherein in the preprocessing module, the superpixel segmentation method comprises the following steps:
initializing a seed point: uniformly distributing seed points in the image according to the set number of the super pixels;
reselecting seed points in n-n neighborhoods of the seed points, calculating gradient values of all pixel points in the neighborhoods, and moving the seed points to the place with the minimum gradient in the neighborhoods, wherein n is 3;
distributing a class label to each pixel point in the neighborhood around each seed point;
distance measurement: the method comprises the steps of calculating the distance between each searched pixel point and the seed point respectively according to the color distance and the space distance;
iterative optimization: continuously iterating the steps until the error is converged, wherein the iteration number is 10;
and (3) enhancing connectivity: and newly building a marking table, wherein the elements in the table are all-1, the discontinuous superpixels and the super-pixels with undersize are redistributed to the adjacent superpixels according to the Z-shaped trend, and the traversed superpixels are distributed to the corresponding labels until all the superpixels are traversed.
3. The method as claimed in claim 1, wherein the extended road shape index extraction module comprises two parts of region extension by taking a segmentation map as a center and spatial feature extraction based on an extended region.
4. The method as claimed in claim 3, wherein the region expansion using the segmentation patches as the center comprises the following steps:
introducing a first geographic law for extracting image spatial features and expressing spatial correlation by using a Moran index;
the method comprises the following steps of (1) carrying out superpixel shape characteristics based on spectral characteristics, spatial characteristics and textural characteristics according to spatial extension characteristics of roads in a high-resolution remote sensing image and uniform regular characteristics of spots segmented by using a SLIC superpixel segmentation method;
and the image spot region expansion algorithm comprises the following steps: inputting a segmented image spot Oc, outputting Or: a set of patches surrounding Or.
5. The method as claimed in claim 4, wherein the blob region expansion algorithm specifically comprises the following steps:
an initialization step: adding Oc to Or;
collecting the pattern spots in contact with the topology in a container Ocon, wherein Ocon is O1, O2, O3, … … Ot;
constructing a feature vector Vc based on the spectrum mean value of the pattern spots Oc, calculating a feature vector Vk of each pattern spot Ocon by the same method, wherein k is more than or equal to 1 and less than or equal to t, comparing the spectrum distance between Vc and Vk, and selecting adjacent pattern spots Os with the nearest distance from Ocon;
comparing Os with Oc, if Os and Oc satisfy constraint rules R1, R2 and R3, Os is accepted as a plaque belonging to the same target source as Oc;
adding Os into Or, and replacing the Oc with Os to perform the next step of extended exploration;
from step 1 to step 5 is an iterative process, and when any of the three constraint rules are not satisfied, the iterative extension terminates and returns Or.
6. The method as claimed in claim 5, wherein the extended region-based spatial feature extraction is performed by: when the expansion iteration around the central patch terminates, a set of uniform and spatially continuous regions Or is output, using the skeleton-based object linearity index and the average road width for describing the spatial features of the regions Or of the patch composition.
7. The method as claimed in claim 1, wherein the road extraction module comprises the following steps:
road extraction based on multi-scale collaborative representation and graph cut algorithm, a multi-scale collaborative representation method is selected, a test sample is represented by linear combination of training samples, the road probability of each object on a certain scale is obtained, all pixels in the same object are endowed with a likelihood value which is the same as that of the object, all road probability graphs of three different scales are fused into a whole, and finally the road possibility of each pixel is obtained;
given an image I, constructing an undirected graph, and obtaining a road probability graph through a multi-scale collaborative representation method so as to segment the image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110914630.XA CN113627331A (en) | 2021-08-10 | 2021-08-10 | High-resolution image road extraction method based on extended road shape index |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110914630.XA CN113627331A (en) | 2021-08-10 | 2021-08-10 | High-resolution image road extraction method based on extended road shape index |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113627331A true CN113627331A (en) | 2021-11-09 |
Family
ID=78384071
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110914630.XA Pending CN113627331A (en) | 2021-08-10 | 2021-08-10 | High-resolution image road extraction method based on extended road shape index |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113627331A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114821333A (en) * | 2022-05-16 | 2022-07-29 | 中国人民解放军61540部队 | High-resolution remote sensing image road material identification method and device |
-
2021
- 2021-08-10 CN CN202110914630.XA patent/CN113627331A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114821333A (en) * | 2022-05-16 | 2022-07-29 | 中国人民解放军61540部队 | High-resolution remote sensing image road material identification method and device |
| CN114821333B (en) * | 2022-05-16 | 2022-11-18 | 中国人民解放军61540部队 | High-resolution remote sensing image road material identification method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhou et al. | SO–CNN based urban functional zone fine division with VHR remote sensing image | |
| Xia et al. | Geometric primitives in LiDAR point clouds: A review | |
| CN107067405B (en) | Remote sensing image segmentation method based on scale optimization | |
| CN111598048B (en) | Urban village-in-village identification method integrating high-resolution remote sensing image and street view image | |
| Liu et al. | Large-scale mapping of gully-affected areas: An approach integrating Google Earth images and terrain skeleton information | |
| CN104408463B (en) | High-resolution construction land pattern spot identification method | |
| CN113223042B (en) | Intelligent acquisition method and equipment for remote sensing image deep learning sample | |
| CN111611960B (en) | Large-area ground surface coverage classification method based on multilayer perceptive neural network | |
| CN113505842A (en) | Automatic extraction method suitable for large-scale regional remote sensing image urban building | |
| CN112084871B (en) | High-resolution remote sensing target boundary extraction method based on weak supervised learning | |
| Qian et al. | Deep Roof Refiner: A detail-oriented deep learning network for refined delineation of roof structure lines using satellite imagery | |
| Truax | Comparing spectral and object based approaches for classification and transportation feature extraction from high resolution multispectral imagery | |
| Muhs et al. | Automatic delineation of built-up area at urban block level from topographic maps | |
| CN113627331A (en) | High-resolution image road extraction method based on extended road shape index | |
| CN111104850A (en) | Remote sensing image building automatic extraction method and system based on residual error network | |
| Zhou et al. | Green urban garden landscape simulation platform based on high-resolution image recognition technology and GIS | |
| CN114140700A (en) | Step-by-step heterogeneous image template matching method based on cascade network | |
| Bashit et al. | Analysis of built-up land spatial patterns using multitemporal satellite imagery in Pekalongan city | |
| CN113837134A (en) | Wetland vegetation identification method based on object-oriented deep learning model and transfer learning | |
| Sha et al. | A supervoxel approach to road boundary enhancement from 3-d lidar point clouds | |
| Lu et al. | Gre and beyond: A global road extraction dataset | |
| Wang et al. | An unsupervised multi-scale segmentation method based on automated parameterization | |
| Dwivedi et al. | New horizons in planning smart cities using LiDAR technology | |
| Forghani et al. | Object-based classification of multi-sensor optical imagery to generate terrain surface roughness information for input to wind risk simulation | |
| Ding et al. | Monitoring and Analysis of Urban Sprawl Based on Road Network Data and High-Resolution Remote Sensing Imagery: A Case Study of China's Provincial Capitals |
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 |