AU2020104112A4 - A RNN based Spatio Temporal Data Mining model for urban road Planning - Google Patents
A RNN based Spatio Temporal Data Mining model for urban road Planning Download PDFInfo
- Publication number
- AU2020104112A4 AU2020104112A4 AU2020104112A AU2020104112A AU2020104112A4 AU 2020104112 A4 AU2020104112 A4 AU 2020104112A4 AU 2020104112 A AU2020104112 A AU 2020104112A AU 2020104112 A AU2020104112 A AU 2020104112A AU 2020104112 A4 AU2020104112 A4 AU 2020104112A4
- Authority
- AU
- Australia
- Prior art keywords
- rnn
- data
- coordinate system
- road
- utm
- 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.)
- Ceased
Links
- 230000002123 temporal effect Effects 0.000 title claims abstract description 18
- 238000007418 data mining Methods 0.000 title claims abstract description 10
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 244000025254 Cannabis sativa Species 0.000 claims abstract description 5
- 239000003086 colorant Substances 0.000 claims abstract description 5
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 5
- 230000010076 replication Effects 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 9
- 230000007246 mechanism Effects 0.000 abstract description 4
- 230000000306 recurrent effect Effects 0.000 description 7
- 238000013459 approach Methods 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000010219 correlation analysis Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000013136 deep learning model Methods 0.000 description 2
- 230000009977 dual effect Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000010422 painting Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000010206 sensitivity analysis Methods 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000012731 temporal analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/38—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system
- G01S19/39—Determining a navigation solution using signals transmitted by a satellite radio beacon positioning system the satellite radio beacon positioning system transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/42—Determining position
- G01S19/51—Relative positioning
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
- G01B21/02—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring length, width, or thickness
- G01B21/04—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring length, width, or thickness by measuring coordinates of points
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C7/00—Tracing profiles
- G01C7/02—Tracing profiles of land surfaces
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/01—Satellite radio beacon positioning systems transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/13—Receivers
- G01S19/14—Receivers specially adapted for specific applications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4046—Scaling of whole images or parts thereof, e.g. expanding or contracting using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Networks & Wireless Communication (AREA)
- Multimedia (AREA)
- Traffic Control Systems (AREA)
Abstract
A RNN BASED SPATIO TEMPORAL DATA MINING MODEL FOR URBAN ROAD
PLANNING
ABSTRACT
As road planning is essential in urban areas, it is necessary to find the mechanisms for
implementing the same. For which first the data need to be extracted. For extracting, the raster
data is used from GPS. The raster data will take the image using GPS. Based on the elevation the
surface was plotted. The different colors as mentioned in figure 1 will categorize the land surfaces
into agriculture, bare ground, water, grass, etc. The surface signified by every cell contains the
same width as well as height. The images are represented in pixels. It is represented in rows and
columns. Then that data needs to be converted into cell (pixel) representation using a cartesian
coordinate system and UTM projected coordinate system. Based on this the east and north
coordinates of the particular region were identified with its position. Now using RNN and spatio
temporal data mining, Directed Acyclic Graph a temporal road using the vector data was built.
Then the shortest path is constructed and the replication was avoided using the Krusal's algorithm.
This results in the way where the road planning was carried out smoothly and efficiently.
1
UTM-Projected
Coordinate System
6,000,000 -r
North
300,000 East 300,700
coordinates are
300,500 E, 5,900,600 N
Figure 5: UTM Projected Coordinate System
Unfold
tw w w w
C I u
f tU tU IU
[out out out2 out
XM x
CC 2 C3
~x x Ix M x
BB B0 B 2 B3
x x x M x
AO A, A2 A,
MX x x Mx
C C C C
in, in, in2 in,
Figure 6: Unfold RNN
3
Description
UTM-Projected Coordinate System
6,000,000 -r
North
300,000 East 300,700
coordinates are 300,500 E, 5,900,600 N
Figure 5: UTM Projected Coordinate System
Unfold tw w w w
CI u f tU tU IU
[out out out2 out XM x
CC 2 C3 ~x x Ix M x B0 BB B B3 2 x x x M x AO A, A2 A, MX x x Mx C C C C
in, in, in2 in,
Figure 6: Unfold RNN
Description
Field of the Invention:
This invention relates to RNN based spatio temporal data mining model for urban road planning. As road planning is essential in urban areas, it is necessary to find the mechanisms for implementing the same. For which first the data need to be extracted. For extracting, the raster data is used from GPS. The raster data will take the image using GPS. Based on the elevation the surface was plotted. The different colors as mentioned in figure 1 will categorize the land surfaces into agriculture, bare ground, water, grass, etc. The surface signified by every cell contains the same width as well as height. The images are represented in pixels. It is represented in rows and columns. Then that data needs to be converted into cell (pixel) representation using a cartesian coordinate system and UTM projected coordinate system. Based on this the east and north coordinates of the region were identified with its position. Now using RNN and spatio temporal data mining, Directed Acyclic Graph a temporal road using the vector data was built. Then the shortest path is constructed and the replication was avoided using the Krusal's algorithm. This results in the way where the road planning was carried out smoothly and efficiently.
Background of the invention:
Senzhang Wang et al. offered a study of emerging raises in the implementation of deep learning methods for (Spatio Temporal Data Mining) STDM. The Spatio-temporal data is classified into five diverse types. Also, bring together the deep learning models that are applied in STDM. Also, it was classified based on the kinds of Spatio-temporal data, the deep learning models, and the data mining tasks trailed by the claims of deep learning for STDM in diverse domains comprising transportation, climate, on-demand service and weather examination, neuroscience, crime analysis, location-based social network, and human motion. In conclusion, they winded up with the limits of current research as well as pinpointed the fore coming research scope.
Wang Xiangxue et al. predicted the traffic flow and framed the same for urban road networks based on data-driven approaches. It is comprised of dual modules. A set of algorithms is comprised of the first module to route traffic flow data. If analysis, as well as repair, get over then a whole data set deprived of outliers is delivered and a data set comprising couples of road segments that are analogous to each other in respect to their fashions. Multiple time-steps short-term forecasting is the second module. Now the decomposition of time series into the trend and residual series takes place with a worthy thoughtful of the periodicity as well as the randomness of traffic flow. Further reconstruction of dual time series, model training as well as forecast based on Long Short Term Memory Recurrent Neural Network (LSTM-RNN) is achieved. In conclusion, the two outcomes are united together to practice the ending prediction. The two urban road networks perform the model evaluation. The consequences show that the data processing module can efficiently advance to reduce the training time, to improve the data quality as well the model robustness also improved. The LSTM-RNN properly classifies the time trend as well as the spatial likeness of traffic flow as well as gets an extra precise numerous time-step prediction. It has better accuracy as well as stability.
Jairo A. G6mez et al. centered on improving three critical community factors, such as demographic distribution, binary urban footprint, and colour urban footprint. With the aid of machine learning, population distribution is modeled as a spatiotemporal regression problem. It thus gains the binary urban footprint from the population distribution through a binary classifier plus a temporal adjustment for predominant urban regions. It also uses the semantic painting algorithm to measure the urban footprint in colour from its earlier meaning, past, and present binary urban footprint values. By entering this network with free data from the Landsat archive as well as the Global Human Settlement Layer framework, attracted users will obtain estimated growth guesses of any city in the world. In the form of supplementary spatially distributed input variables concluded with time emphasis to availability, these calculations can be improved with the insertion. Through the projected machine learning-based architecture built on cellular automata, two advantages vary from commonly used growth models. Essentially, because the model studies the dynamics of growth directly from historical data, it is not necessary to express rules a priori. Also, to assess their effect, it is simple to train new machine learning models using various instructive input variables. Implementation in real-time has taken place in Valledupar and Rionegro. They are the two Latin American cities with different geomorphological features located in Colombia, as well as the model guesses based on output metrics are in adjacent arrangement with the ground-truth. Pearson's correlation coefficient for continuous variables, the root-mean-square error, the output measures that are known to be zero-mean normalized cross-correlation. For discrete variables such as intersection above union, the fl metric, and precision, others are also uncommon. In conclusion, the methodology is flexible for urban growth modeling. It allows sensitivity analysis and enables policymakers around the world to assess different what-if scenarios in resilient cities.
Zheyi Pan et al., using smart transport networks as well as public safety, forecasts urban traffic. It is challenging based on two aspects: 1) dynamic Spatio-temporal urban traffic correlations; 2) a variety of spatiotemporal correlations. A deep meta-learning based model, abbreviated as ST MetaNet, was proposed to solve these challenges. It helps instantly predict traffic in any area. ST MetaNet uses an encoder to research historical data as well as a periodic prediction decoder. The encoder and the decoder have the same network structure consisting of a recurring neural traffic encoding network. To arrest various spatial correlations, the Meta graph attention network is used as well as the Meta recurrent neural network to ruminate different temporal correlations. Extensive experiments have contributed to the efficiency of ST-MetaNet based on two real-world datasets to demonstrate various methods.
Guowen Dai et al. clarified the prediction model for traffic flow that pooled the Spatio-temporal analysis with a Gated Recurrent Unit (GRU). Based on the composite traffic flow results primarily time correlation analysis as well as spatial correlation analysis was achieved. To define the optimal input time interval and spatial data volume, the spatiotemporal feature selection algorithm was also employed. The traffic flow data were also derived from the existing traffic flow data and translated into a two-dimensional matrix. The GRU was used to practice the Spatio-temporal data function of the matrix's internal traffic flow to achieve forecast determination. In conclusion, to prove the efficiency of the model, the prediction fallouts acquired by the projected model were correlated with the real traffic flow results. The suggested approach outdoes both accuracy and stability.
Spatio Temporal Graph Convolutional Networks (STGCN) was proposed by Bing-Yu et al. to challenge the time series to think tricky in the traffic domain. The difficulty in graphs as well as building the model with entire convolutional structures was expressed in its place of relating normal convolutional as well as recurrent units, enabling faster training speed with fewer parameters. Trials show that the STGCN model effectively arrests full Spatio-temporal associations over multi-scale traffic network modeling and beats baselines on multiple real-world traffic datasets reliably.
The approach that is suitable for grid-based Spatio-temporal prediction in condensed urban areas was proposed by Yang Liu et al. Research shows that many base models of traffic state prediction can be pooled to advance the accuracy of the prediction.
Pavlyuk et al. proposed a spatiotemporal method that instantaneously exploits both spatial as well as temporal relationships in the arena of traffic flow forecasting. Precise proof of identity of the spatiotemporal construction shows a precarious part in current traffic forecasting procedures as well as fresh advances of data-driven feature variety as well as extraction approaches permit the proof of identity of complex relations. It methodically analyses studies that apply feature selection as well as abstraction approaches for spatiotemporal traffic forecasting. A blend of bibliographic sources illuminates the benefits and weaknesses of diverse feature selection as well as extraction methods for culture the spatiotemporal structure as well as notices fashions in their submissions.
Ya Zhang et al., suggest that traffic forecasting is a significant precondition for the demand for intelligent transportation systems in urban traffic networks. To apply GCRN to the large-scale road networks based on great computational complexity is a challenge. They proposed theorizing the road network into a geometric graph. A Fast Graph Convolution Recurrent Neural Network (FastGCRNN) is constructed to model the spatial-temporal dependencies of traffic flow. Exactly FastGCN unit is effective to arrest the topological association among the roads as well as the adjacent roads in the graph with dropping the computational complexity over importance sampling. Pooling GRU unit to arrest the temporal dependency of traffic flow as well as surround the spatiotemporal features into Seq2Seq created on the Encoder-Decoder framework. Investigations on large scale traffic data sets exemplify that the projected method significantly decreases computational complexity as well as memory consumption in preserving moderately great accuracy.
Huaxia Yao et al. made two vital explanations: (1) the spatial dependencies among locations are dynamic (2) the temporal dependency trails daily as well as weekly pattern and not periodic for its dynamic temporal shifting. To remedy these two problems Spatial-Temporal Dynamic Network (STDN) is proposed. Flow gating mechanism is familiarized to study the dynamic resemblance among locations as well as occasionally shifted attention mechanism is planned to lever long-term periodic temporal shifting. This results in verifying the effectiveness of real-world traffic datasets.
Objects of the Invention:
• To extract raster data from GPS. • To convert the raster data into cell(pixel) representation using a cartesian coordinate system and UTM projected coordinate system • Based on RNN find the Directed Acyclic Graph and make a temporal road using the vector data arrived with. • Using Kruskal's algorithm, find the shortest route from the derived spanning tree • Also, avoid the replication using the Krusal's algorithm
Summary of the Invention
In urban areas, road planning is an essential one. For which the data which is there needs to be considered. Raster data is an essential type which is useful for projecting the transportation. The raster data will capture the image using (Global Positioning System) GPS. Based on the elevation the surface was plotted. Also, the different colors as mentioned in figure 1 will categorize the land surfaces into agriculture, bare ground, water, grass, etc.
The surface signified by every cell contains the same width as well as height. Also, it is an identical share of the whole surface signified by the raster. The images are represented in pixels. The pixel format is depicted in figure 2. The area or surface was represented in cells. Based on the coverage it is accommodated. It is represented in rows and columns. The cells are represented using the Cartesian coordinate system as in figure 4. After converting into the Cartesian representation the same was associated with the Universal Transverse Mercator (UTM) projected coordinate system. Based on this the east and north coordinates of the particular region were identified with its position as in figure 5.
Further by getting the coordinates, it is plotted as a graph using STDM. As in (Recurrent Neural Network) RNN there are two types of connection named as fold RNN and unfold RNN, unfold RNN is chosen here to carry over the process. The same was depicted in figure 6. A Directed Acyclic Graph (DAG) was constructed. The feed forwarding mechanism was implemented.
Now based on the graph connected with nodes and vertices without any cycle helps to find the routes using Kruskal's spanning tree algorithm. As in Kruskal's algorithm, it will be taking all possible spanning trees, and also not having cycles is considered so it was chosen to plan the road. By using so the distance between the vertices are identified and it was allotted as weights in the nodes. Based on the weights the road are constructed in the urban areas
Detailed Description of the Invention:
In urban areas, road planning is an essential one. For which the data which is there needs to be considered. Raster data is an essential type which is useful for projecting the transportation. The raster data will capture the image using GPS. Based on the elevation the surface was plotted. Also, the different colours as mentioned in figure 1 will categorize the land surfaces into agriculture, bare ground, water, grass, etc.
The surface signified by every cell contains the same width as well as height. Also, it is an identical share of the whole surface signified by the raster. The images are represented in pixels. The pixel format is depicted in figure 2. The area or surface was represented in cells. Based on the coverage it is accommodated. It is represented in rows and columns. The cells are represented using the Cartesian coordinate system as in figure 4. After converting into the Cartesian representation the same was associated with the Universal Transverse Mercator (UTM) projected coordinate system. Based on this the east and north coordinates of the particular region were identified with its position as in figure 5.
Further by getting the coordinates, it is plotted as a graph using STDM. As in (Recurrent Neural Network) RNN there are two types of connection named as fold RNN and unfold RNN, unfold RNN is chosen here to carry over the process. The same was depicted in figure 6. A Directed Acyclic Graph (DAG) was constructed. The feed forwarding mechanism was implemented.
Now based on the graph connected with nodes and vertices without any cycle helps to find the routes using Kruskal's spanning tree algorithm. As in Kruskal's algorithm, it will be taking all possible spanning trees, and also not having cycles is considered so it was chosen to plan the road. By using so the distance between the vertices are identified and it was allotted as weights in the nodes. Based on the weights the road are constructed in the urban areas
Claims (4)
1. Extracting raster data from GPS. a. The raster data will take the image using GPS. b. Based on the elevation the surface was plotted. c. The different colors as mentioned in figure 1 will categorize the land surfaces into agriculture, bare ground, water, grass, etc. d. The surface signified by every cell contains the same width as well as height. e. The images are represented in pixels. f. It is represented in rows and columns.
2. Converting the raster data into cell(pixel) representation using a cartesian coordinate system and UTM projected coordinate system a. Based on this the east and north coordinates of the region was identified with its position
3. Creation of Directed Acyclic Graph a temporal road using the vector data
4. Deriving shortest path and to avoid the replication using the Krusal's algorithm
A RNN BASED SPATIO TEMPORAL DATA MINING MODEL FOR URBAN ROAD PLANNING
Drawings 2020104112
Fig 1: Categorization of surface
Figure 2: Pixel representation
Figure 3: Representation of surface in the cell
Figure 4: Cartesian Coordinate System
Figure 5: UTM Projected Coordinate System
Figure 6: Unfold RNN
Figure 7: Planning road vector using RNN
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU2020104112A AU2020104112A4 (en) | 2020-12-15 | 2020-12-15 | A RNN based Spatio Temporal Data Mining model for urban road Planning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
AU2020104112A AU2020104112A4 (en) | 2020-12-15 | 2020-12-15 | A RNN based Spatio Temporal Data Mining model for urban road Planning |
Publications (1)
Publication Number | Publication Date |
---|---|
AU2020104112A4 true AU2020104112A4 (en) | 2021-03-04 |
Family
ID=74715979
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
AU2020104112A Ceased AU2020104112A4 (en) | 2020-12-15 | 2020-12-15 | A RNN based Spatio Temporal Data Mining model for urban road Planning |
Country Status (1)
Country | Link |
---|---|
AU (1) | AU2020104112A4 (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113051811A (en) * | 2021-03-16 | 2021-06-29 | 重庆邮电大学 | Multi-mode short-term traffic jam prediction method based on GRU network |
CN117407477A (en) * | 2023-10-26 | 2024-01-16 | 航科院中宇(北京)新技术发展有限公司 | Geographic information data evolution recognition processing method, system and storage medium |
-
2020
- 2020-12-15 AU AU2020104112A patent/AU2020104112A4/en not_active Ceased
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113051811A (en) * | 2021-03-16 | 2021-06-29 | 重庆邮电大学 | Multi-mode short-term traffic jam prediction method based on GRU network |
CN113051811B (en) * | 2021-03-16 | 2022-08-05 | 重庆邮电大学 | Multi-mode short-term traffic jam prediction method based on GRU network |
CN117407477A (en) * | 2023-10-26 | 2024-01-16 | 航科院中宇(北京)新技术发展有限公司 | Geographic information data evolution recognition processing method, system and storage medium |
CN117407477B (en) * | 2023-10-26 | 2024-05-14 | 航科院中宇(北京)新技术发展有限公司 | Geographic information data evolution recognition processing method, system and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liu et al. | STMGCN: Mobile edge computing-empowered vessel trajectory prediction using spatio-temporal multigraph convolutional network | |
Felemban et al. | Digital revolution for Hajj crowd management: A technology survey | |
AU2020104112A4 (en) | A RNN based Spatio Temporal Data Mining model for urban road Planning | |
Mohammady et al. | Urban sprawl assessment and modeling using landsat images and GIS | |
Niu et al. | Real-time taxi-passenger prediction with L-CNN | |
Saxena et al. | D-GAN: Deep generative adversarial nets for spatio-temporal prediction | |
CN111611949A (en) | Intelligent urban water area garbage density detection method based on CIM and block chain | |
Tran et al. | Pp-linknet: Improving semantic segmentation of high resolution satellite imagery with multi-stage training | |
Gao et al. | Road extraction using a dual attention dilated-linknet based on satellite images and floating vehicle trajectory data | |
Bapaume et al. | Image inpainting and deep learning to forecast short-term train loads | |
CN111627103A (en) | Smart city CIM imaging method based on pedestrian activity and density perception | |
CN113159371B (en) | Unknown target feature modeling and demand prediction method based on cross-modal data fusion | |
Stark | Using deep convolutional neural networks for the identification of informal settlements to improve a sustainable development in urban environments | |
Hetang et al. | Segment Anything Model for Road Network Graph Extraction | |
Hermes et al. | A graph-based U-net model for predicting traffic in unseen cities | |
Nguyen et al. | Multi-task learning for maritime traffic surveillance from AIS data streams | |
Hu et al. | A spatiotemporal intelligent framework and experimental platform for urban digital twins | |
CN114419338B (en) | Image processing method, image processing device, computer equipment and storage medium | |
Değermenci | Spatio-temporal change analysis and prediction of land use and land cover changes using CA-ANN model | |
Sujiwo et al. | Robust and accurate monocular vision-based localization in outdoor environments of real-world robot challenge | |
Tsanakas et al. | A light-weight edge-enabled knowledge distillation technique for next location prediction of multitude transportation means | |
Chen et al. | Deep learning with satellite images and volunteered geographic information | |
KR102295671B1 (en) | Method for predicting human mobility route based on a generative adversarial network | |
Tang et al. | Discovery of migration habitats and routes of wild bird species by clustering and association analysis | |
Hu et al. | Road extraction from satellite image via auxiliary road location prediction |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
FGI | Letters patent sealed or granted (innovation patent) | ||
MK22 | Patent ceased section 143a(d), or expired - non payment of renewal fee or expiry |