1
|
Gonçalves DN, Junior JM, Arruda MDSD, Fernandes VJM, Ramos APM, Furuya DEG, Osco LP, He H, Jorge LADC, Li J, Melgani F, Pistori H, Gonçalves WN. A deep learning approach based on graphs to detect plantation lines. Heliyon 2024; 10:e31730. [PMID: 38841473 PMCID: PMC11152659 DOI: 10.1016/j.heliyon.2024.e31730] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/22/2023] [Revised: 05/21/2024] [Accepted: 05/21/2024] [Indexed: 06/07/2024] Open
Abstract
Identifying plantation lines in aerial images of agricultural landscapes is re-quired for many automatic farming processes. Deep learning-based networks are among the most prominent methods to learn such patterns and extract this type of information from diverse imagery conditions. However, even state-of-the-art methods may stumble in complex plantation patterns. Here, we propose a deep learning approach based on graphs to detect plantation lines in UAV-based RGB imagery, presenting a challenging scenario containing spaced plants. The first module of our method extracts a feature map throughout the backbone, which consists of the initial layers of the VGG16. This feature map is used as an input to the Knowledge Estimation Module (KEM), organized in three concatenated branches for detecting 1) the plant positions, 2) the plantation lines, and 3) the displacement vectors between the plants. A graph modeling is applied considering each plant position on the image as vertices, and edges are formed between two vertices (i.e. plants). Finally, the edge is classified as pertaining to a certain plantation line based on three probabilities (higher than 0.5): i) in visual features obtained from the backbone; ii) a chance that the edge pixels belong to a line, from the KEM step; and iii) an alignment of the displacement vectors with the edge, also from the KEM step. Experiments were conducted initially in corn plantations with different growth stages and patterns with aerial RGB imagery to present the advantages of adopting each module. We assessed the generalization capability in the other two cultures (orange and eucalyptus) datasets. The proposed method was compared against state-of-the-art deep learning methods and achieved superior performance with a significant margin considering all three datasets. This approach is useful in extracting lines with spaced plantation patterns and could be implemented in scenarios where plantation gaps occur, generating lines with few-to-no interruptions.
Collapse
Affiliation(s)
- Diogo Nunes Gonçalves
- Faculty of Computer Science, Federal University of Mato Grosso do Sul, Av. Costa e Silva, Campo Grande, 79070-900, MS, Brazil
| | - José Marcato Junior
- Faculty of Engineering, Architecture, and Urbanism and Geography, Federal University of Mato Grosso do Sul, Av. Costa e Silva, Campo Grande, 79070-900, MS, Brazil
| | - Mauro dos Santos de Arruda
- Faculty of Computer Science, Federal University of Mato Grosso do Sul, Av. Costa e Silva, Campo Grande, 79070-900, MS, Brazil
| | | | - Ana Paula Marques Ramos
- Faculty of Science and Technology, São Paulo State University (UNESP), R. Roberto Simonsen, 305, Presidente Prudente 19060-900, SP, Brazil
| | - Danielle Elis Garcia Furuya
- Program of Environment and Regional Developement, University of Western São Paulo, Raposo Tavares, km 572, Presidente Prudente, 19067-175, SP, Brazil
| | - Lucas Prado Osco
- Program of Environment and Regional Developement, University of Western São Paulo, Raposo Tavares, km 572, Presidente Prudente, 19067-175, SP, Brazil
| | - Hongjie He
- Department of Geography and Environmental Management, University of Waterloo, Waterloo, ON N2L 3G1, Canada
| | - Lucio André de Castro Jorge
- National Research Center of Development of Agricultural Instrumentation, Brazilian Agricultural Research Agency (EMBRAPA), 13560-970, R. XV de Novembro, 1452, São Carlos, SP, Brazil
| | - Jonathan Li
- Department of Geography and Environmental Management, University of Waterloo, Waterloo, ON N2L 3G1, Canada
| | - Farid Melgani
- Department of Information Engineering and Computer Science, University of Trento, Trento, 38122, Italy
| | - Hemerson Pistori
- Faculty of Computer Science, Federal University of Mato Grosso do Sul, Av. Costa e Silva, Campo Grande, 79070-900, MS, Brazil
- INOVISAO, Dom Bosco Catholic University, Avenida Tamandaré, 6000, Campo Grande, 79117-900, MS, Brazil
| | - Wesley Nunes Gonçalves
- Faculty of Computer Science, Federal University of Mato Grosso do Sul, Av. Costa e Silva, Campo Grande, 79070-900, MS, Brazil
- Faculty of Engineering, Architecture, and Urbanism and Geography, Federal University of Mato Grosso do Sul, Av. Costa e Silva, Campo Grande, 79070-900, MS, Brazil
| |
Collapse
|
2
|
Jain A, Laidlaw DH, Bajcsy P, Singh R. Memory-efficient semantic segmentation of large microscopy images using graph-based neural networks. Microscopy (Oxf) 2024; 73:275-286. [PMID: 37864808 DOI: 10.1093/jmicro/dfad049] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/13/2023] [Revised: 07/14/2023] [Accepted: 10/05/2023] [Indexed: 10/23/2023] Open
Abstract
We present a graph neural network (GNN)-based framework applied to large-scale microscopy image segmentation tasks. While deep learning models, like convolutional neural networks (CNNs), have become common for automating image segmentation tasks, they are limited by the image size that can fit in the memory of computational hardware. In a GNN framework, large-scale images are converted into graphs using superpixels (regions of pixels with similar color/intensity values), allowing us to input information from the entire image into the model. By converting images with hundreds of millions of pixels to graphs with thousands of nodes, we can segment large images using memory-limited computational resources. We compare the performance of GNN- and CNN-based segmentation in terms of accuracy, training time and required graphics processing unit memory. Based on our experiments with microscopy images of biological cells and cell colonies, GNN-based segmentation used one to three orders-of-magnitude fewer computational resources with only a change in accuracy of ‒2 % to +0.3 %. Furthermore, errors due to superpixel generation can be reduced by either using better superpixel generation algorithms or increasing the number of superpixels, thereby allowing for improvement in the GNN framework's accuracy. This trade-off between accuracy and computational cost over CNN models makes the GNN framework attractive for many large-scale microscopy image segmentation tasks in biology.
Collapse
Affiliation(s)
- Atishay Jain
- Department of Computer Science, Brown University, 115 Waterman Street, Providence, Rhode Island 02906, USA
| | - David H Laidlaw
- Department of Computer Science, Brown University, 115 Waterman Street, Providence, Rhode Island 02906, USA
| | - Peter Bajcsy
- Information Technology Laboratory, National Institute of Standards and Technology (NIST), 100 Bureau Drive, Gaithersburg, Maryland 20899, USA
| | - Ritambhara Singh
- Department of Computer Science, Brown University, 115 Waterman Street, Providence, Rhode Island 02906, USA
- Center for Computational Molecular Biology, Brown University, 164 Angell Street, Providence, Rhode Island 02906, USA
| |
Collapse
|
3
|
Turčinović F, Kačan M, Bojanjac D, Bosiljevac M, Šipuš Z. Utilizing Polarization Diversity in GBSAR Data-Based Object Classification. SENSORS (BASEL, SWITZERLAND) 2024; 24:2305. [PMID: 38610516 PMCID: PMC11014032 DOI: 10.3390/s24072305] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 02/12/2024] [Revised: 03/28/2024] [Accepted: 04/03/2024] [Indexed: 04/14/2024]
Abstract
In recent years, the development of intelligent sensor systems has experienced remarkable growth, particularly in the domain of microwave and millimeter wave sensing, thanks to the increased availability of affordable hardware components. With the development of smart Ground-Based Synthetic Aperture Radar (GBSAR) system called GBSAR-Pi, we previously explored object classification applications based on raw radar data. Building upon this foundation, in this study, we analyze the potential of utilizing polarization information to improve the performance of deep learning models based on raw GBSAR data. The data are obtained with a GBSAR operating at 24 GHz with both vertical (VV) and horizontal (HH) polarization, resulting in two matrices (VV and HH) per observed scene. We present several approaches demonstrating the integration of such data into classification models based on a modified ResNet18 architecture. We also introduce a novel Siamese architecture tailored to accommodate the dual input radar data. The results indicate that a simple concatenation method is the most promising approach and underscore the importance of considering antenna polarization and merging strategies in deep learning applications based on radar data.
Collapse
Affiliation(s)
| | | | | | | | - Zvonimir Šipuš
- Faculty of Electrical Engineering and Computing, University of Zagreb, 10 000 Zagreb, Croatia; (F.T.); (M.K.); (D.B.); (M.B.)
| |
Collapse
|
4
|
Orozco J, Manian V, Alfaro E, Walia H, Dhatt BK. Graph Convolutional Network Using Adaptive Neighborhood Laplacian Matrix for Hyperspectral Images with Application to Rice Seed Image Classification. SENSORS (BASEL, SWITZERLAND) 2023; 23:3515. [PMID: 37050573 PMCID: PMC10099153 DOI: 10.3390/s23073515] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Subscribe] [Scholar Register] [Received: 02/25/2023] [Revised: 03/21/2023] [Accepted: 03/25/2023] [Indexed: 06/19/2023]
Abstract
Graph convolutional neural network architectures combine feature extraction and convolutional layers for hyperspectral image classification. An adaptive neighborhood aggregation method based on statistical variance integrating the spatial information along with the spectral signature of the pixels is proposed for improving graph convolutional network classification of hyperspectral images. The spatial-spectral information is integrated into the adjacency matrix and processed by a single-layer graph convolutional network. The algorithm employs an adaptive neighborhood selection criteria conditioned by the class it belongs to. Compared to fixed window-based feature extraction, this method proves effective in capturing the spectral and spatial features with variable pixel neighborhood sizes. The experimental results from the Indian Pines, Houston University, and Botswana Hyperion hyperspectral image datasets show that the proposed AN-GCN can significantly improve classification accuracy. For example, the overall accuracy for Houston University data increases from 81.71% (MiniGCN) to 97.88% (AN-GCN). Furthermore, the AN-GCN can classify hyperspectral images of rice seeds exposed to high day and night temperatures, proving its efficacy in discriminating the seeds under increased ambient temperature treatments.
Collapse
Affiliation(s)
- Jairo Orozco
- University of Puerto Rico at Mayaguez, Mayagüez, PR 00681, USA
| | - Vidya Manian
- University of Puerto Rico at Mayaguez, Mayagüez, PR 00681, USA
| | | | | | | |
Collapse
|
5
|
Li S, Liu F, Jiao L, Chen P, Liu X, Li L. MFNet: A Novel GNN-Based Multi-Level Feature Network With Superpixel Priors. IEEE TRANSACTIONS ON IMAGE PROCESSING : A PUBLICATION OF THE IEEE SIGNAL PROCESSING SOCIETY 2022; 31:7306-7321. [PMID: 36383578 DOI: 10.1109/tip.2022.3220057] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/16/2023]
Abstract
Since the superpixel segmentation method aggregates pixels based on similarity, the boundaries of some superpixels indicate the outline of the object and the superpixels provide prerequisites for learning structural-aware features. It is worthwhile to research how to utilize these superpixel priors effectively. In this work, by constructing the graph within superpixel and the graph among superpixels, we propose a novel Multi-level Feature Network (MFNet) based on graph neural network with the above superpixel priors. In our MFNet, we learn three-level features in a hierarchical way: from pixel-level feature to superpixel-level feature, and then to image-level feature. To solve the problem that the existing methods cannot represent superpixels well, we propose a superpixel representation method based on graph neural network, which takes the graph constructed by a single superpixel as input to extract the feature of the superpixel. To reflect the versatility of our MFNet, we apply it to an image-level prediction task and a pixel-level prediction task by designing different prediction modules. An attention linear classifier prediction module is proposed for image-level prediction tasks, such as image classification. An FC-based superpixel prediction module and a Decoder-based pixel prediction module are proposed for pixel-level prediction tasks, such as salient object detection. Our MFNet achieves competitive results on a number of datasets when compared with related methods. The visualization shows that the object boundaries and outline of the saliency maps predicted by our proposed MFNet are more refined and pay more attention to details.
Collapse
|
6
|
3T-IEC*: a context-aware recommender system architecture for smart social networks (EBSN and SBSN). J Intell Inf Syst 2022. [DOI: 10.1007/s10844-022-00743-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/10/2022]
|
7
|
Liu F, Qian X, Jiao L, Zhang X, Li L, Cui Y. Contrastive Learning-Based Dual Dynamic GCN for SAR Image Scene Classification. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2022; PP:390-404. [PMID: 35594238 DOI: 10.1109/tnnls.2022.3174873] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 06/15/2023]
Abstract
As a typical label-limited task, it is significant and valuable to explore networks that enable to utilize labeled and unlabeled samples simultaneously for synthetic aperture radar (SAR) image scene classification. Graph convolutional network (GCN) is a powerful semisupervised learning paradigm that helps to capture the topological relationships of scenes in SAR images. While the performance is not satisfactory when existing GCNs are directly used for SAR image scene classification with limited labels, because few methods to characterize the nodes and edges for SAR images. To tackle these issues, we propose a contrastive learning-based dual dynamic GCN (DDGCN) for SAR image scene classification. Specifically, we design a novel contrastive loss to capture the structures of views and scenes, and develop a clustering-based contrastive self-supervised learning model for mapping SAR images from pixel space to high-level embedding space, which facilitates the subsequent node representation and message passing in GCNs. Afterward, we propose a multiple features and parameter sharing dual network framework called DDGCN. One network is a dynamic GCN to keep the local consistency and nonlocal dependency of the same scene with the help of a node attention module and a dynamic correlation matrix learning algorithm. The other is a multiscale and multidirectional fully connected network (FCN) to enlarge the discrepancies between different scenes. Finally, the features obtained by the two branches are fused for classification. A series of experiments on synthetic and real SAR images demonstrate that the proposed method achieves consistently better classification performance than the existing methods.
Collapse
|
8
|
Effect of Attention Mechanism in Deep Learning-Based Remote Sensing Image Processing: A Systematic Literature Review. REMOTE SENSING 2021. [DOI: 10.3390/rs13152965] [Citation(s) in RCA: 21] [Impact Index Per Article: 7.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 12/12/2022]
Abstract
Machine learning, particularly deep learning (DL), has become a central and state-of-the-art method for several computer vision applications and remote sensing (RS) image processing. Researchers are continually trying to improve the performance of the DL methods by developing new architectural designs of the networks and/or developing new techniques, such as attention mechanisms. Since the attention mechanism has been proposed, regardless of its type, it has been increasingly used for diverse RS applications to improve the performances of the existing DL methods. However, these methods are scattered over different studies impeding the selection and application of the feasible approaches. This study provides an overview of the developed attention mechanisms and how to integrate them with different deep learning neural network architectures. In addition, it aims to investigate the effect of the attention mechanism on deep learning-based RS image processing. We identified and analyzed the advances in the corresponding attention mechanism-based deep learning (At-DL) methods. A systematic literature review was performed to identify the trends in publications, publishers, improved DL methods, data types used, attention types used, overall accuracies achieved using At-DL methods, and extracted the current research directions, weaknesses, and open problems to provide insights and recommendations for future studies. For this, five main research questions were formulated to extract the required data and information from the literature. Furthermore, we categorized the papers regarding the addressed RS image processing tasks (e.g., image classification, object detection, and change detection) and discussed the results within each group. In total, 270 papers were retrieved, of which 176 papers were selected according to the defined exclusion criteria for further analysis and detailed review. The results reveal that most of the papers reported an increase in overall accuracy when using the attention mechanism within the DL methods for image classification, image segmentation, change detection, and object detection using remote sensing images.
Collapse
|
9
|
Cui W, Yao M, Hao Y, Wang Z, He X, Wu W, Li J, Zhao H, Xia C, Wang J. Knowledge and Geo-Object Based Graph Convolutional Network for Remote Sensing Semantic Segmentation. SENSORS 2021; 21:s21113848. [PMID: 34199626 PMCID: PMC8199747 DOI: 10.3390/s21113848] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 03/06/2021] [Revised: 05/05/2021] [Accepted: 05/30/2021] [Indexed: 11/16/2022]
Abstract
Pixel-based semantic segmentation models fail to effectively express geographic objects and their topological relationships. Therefore, in semantic segmentation of remote sensing images, these models fail to avoid salt-and-pepper effects and cannot achieve high accuracy either. To solve these problems, object-based models such as graph neural networks (GNNs) are considered. However, traditional GNNs directly use similarity or spatial correlations between nodes to aggregate nodes’ information, which rely too much on the contextual information of the sample. The contextual information of the sample is often distorted, which results in a reduction in the node classification accuracy. To solve this problem, a knowledge and geo-object-based graph convolutional network (KGGCN) is proposed. The KGGCN uses superpixel blocks as nodes of the graph network and combines prior knowledge with spatial correlations during information aggregation. By incorporating the prior knowledge obtained from all samples of the study area, the receptive field of the node is extended from its sample context to the study area. Thus, the distortion of the sample context is overcome effectively. Experiments demonstrate that our model is improved by 3.7% compared with the baseline model named Cluster GCN and 4.1% compared with U-Net.
Collapse
Affiliation(s)
- Wei Cui
- Correspondence: ; Tel.: +86-136-2860-8563
| | | | | | | | | | | | | | | | | | | |
Collapse
|
10
|
Knowledge and Spatial Pyramid Distance-Based Gated Graph Attention Network for Remote Sensing Semantic Segmentation. REMOTE SENSING 2021. [DOI: 10.3390/rs13071312] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
The pixel-based semantic segmentation methods take pixels as recognitions units, and are restricted by the limited range of receptive fields, so they cannot carry richer and higher-level semantics. These reduce the accuracy of remote sensing (RS) semantic segmentation to a certain extent. Comparing with the pixel-based methods, the graph neural networks (GNNs) usually use objects as input nodes, so they not only have relatively small computational complexity, but also can carry richer semantic information. However, the traditional GNNs are more rely on the context information of the individual samples and lack geographic prior knowledge that reflects the overall situation of the research area. Therefore, these methods may be disturbed by the confusion of “different objects with the same spectrum” or “violating the first law of geography” in some areas. To address the above problems, we propose a remote sensing semantic segmentation model called knowledge and spatial pyramid distance-based gated graph attention network (KSPGAT), which is based on prior knowledge, spatial pyramid distance and a graph attention network (GAT) with gating mechanism. The model first uses superpixels (geographical objects) to form the nodes of a graph neural network and then uses a novel spatial pyramid distance recognition algorithm to recognize the spatial relationships. Finally, based on the integration of feature similarity and the spatial relationships of geographic objects, a multi-source attention mechanism and gating mechanism are designed to control the process of node aggregation, as a result, the high-level semantics, spatial relationships and prior knowledge can be introduced into a remote sensing semantic segmentation network. The experimental results show that our model improves the overall accuracy by 4.43% compared with the U-Net Network, and 3.80% compared with the baseline GAT network.
Collapse
|
11
|
Object Detection and Image Segmentation with Deep Learning on Earth Observation Data: A Review—Part II: Applications. REMOTE SENSING 2020. [DOI: 10.3390/rs12183053] [Citation(s) in RCA: 61] [Impact Index Per Article: 15.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 12/17/2022]
Abstract
In Earth observation (EO), large-scale land-surface dynamics are traditionally analyzed by investigating aggregated classes. The increase in data with a very high spatial resolution enables investigations on a fine-grained feature level which can help us to better understand the dynamics of land surfaces by taking object dynamics into account. To extract fine-grained features and objects, the most popular deep-learning model for image analysis is commonly used: the convolutional neural network (CNN). In this review, we provide a comprehensive overview of the impact of deep learning on EO applications by reviewing 429 studies on image segmentation and object detection with CNNs. We extensively examine the spatial distribution of study sites, employed sensors, used datasets and CNN architectures, and give a thorough overview of applications in EO which used CNNs. Our main finding is that CNNs are in an advanced transition phase from computer vision to EO. Upon this, we argue that in the near future, investigations which analyze object dynamics with CNNs will have a significant impact on EO research. With a focus on EO applications in this Part II, we complete the methodological review provided in Part I.
Collapse
|
12
|
DeepInSAR—A Deep Learning Framework for SAR Interferometric Phase Restoration and Coherence Estimation. REMOTE SENSING 2020. [DOI: 10.3390/rs12142340] [Citation(s) in RCA: 12] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
Over the past decade, using Interferometric Synthetic Aperture Radar (InSAR) remote sensing technology for ground displacement detection has become very successful. However, during the acquisition stage, microwave signals reflected from the ground and received by the satellite are contaminated, for example, due to undesirable material reflectance and atmospheric factors, and there is no clean ground truth to discriminate these noises, which adversely affect InSAR phase computation. Accurate InSAR phase filtering and coherence estimation are crucial for subsequent processing steps. Current methods require expert supervision and expensive runtime to evaluate the quality of intermediate outputs, limiting the usability and scalability in practical applications, such as wide area ground displacement monitoring and predication. We propose a deep convolutional neural network based model DeepInSAR to intelligently solve both phase filtering and coherence estimation problems. We demonstrate our model’s performance using simulated and real data. A teacher-student framework is introduced to handle the issue of missing clean InSAR ground truth. Quantitative and qualitative evaluations show that our teacher-student approach requires less input but can achieve better results than its stack-based teacher method even on new unseen data. The proposed DeepInSAR also outperforms three other top non-stack based methods in time efficiency without human supervision.
Collapse
|
13
|
Abstract
The traditional unsupervised image segmentation methods are widely used in synthetic aperture radar (SAR) image segmentation due to the simple and convenient application process. In order to solve the time-consuming problem of the common methods, an SAR image segmentation method using region smoothing and label correction (RSLC) is proposed. In this algorithm, the image smoothing results are used to approximate the results of the spatial information polynomials of the image. Thus, the segmentation process can be realized quickly and effectively. Firstly, direction templates are used to detect the directions at different coordinates of the image, and smoothing templates are used to smooth the edge regions according to the directions. It achieves the smoothing of the edge regions and the retention of the edge information. Then the homogeneous regions are presented indirectly according to the difference of directions. The homogeneous regions are smoothed by using isotropic operators. Finally, the two regions are fused for K-means clustering. The majority voting algorithm is used to modify the clustering results, and the final segmentation results are obtained. Experimental results on simulated SAR images and real SAR images show that the proposed algorithm outperforms the other five state-of-the-art algorithms in segmentation speed and accuracy.
Collapse
|
14
|
A Spatial Adaptive Algorithm Framework for Building Pattern Recognition Using Graph Convolutional Networks. SENSORS 2019; 19:s19245518. [PMID: 31847218 PMCID: PMC6960836 DOI: 10.3390/s19245518] [Citation(s) in RCA: 10] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Subscribe] [Scholar Register] [Received: 10/19/2019] [Revised: 12/08/2019] [Accepted: 12/11/2019] [Indexed: 12/29/2022]
Abstract
Graph learning methods, especially graph convolutional networks, have been investigated for their potential applicability in many fields of study based on topological data. Their topological data processing capabilities have proven to be powerful. However, the relationships among separate entities include not only topological adjacency, but also correlation in vision, for example, the spatial vector data of buildings. In this study, we propose a spatial adaptive algorithm framework with a data-driven design to accomplish building group division and building group pattern recognition tasks, which is not sensitive to the difference in the spatial distribution of the buildings in various geographical regions. In addition, the algorithm framework has a multi-stage design, and processes the building group data from whole to parts, since the objective is closely related to multi-object detection on topological data. By using the graph convolution method and a deep neural network (DNN), the multitask model in this study can learn human thoughts through supervised training, and the whole process only depends upon the descriptive vector data of buildings without any ancillary data for building group partition. Experiments confirmed that the method for expressing buildings and the effect of the algorithm framework proposed are satisfactory. In summary, using deep learning methods to complete the tasks of building group division and building group pattern recognition is potentially effective, and the algorithm framework is worth further research.
Collapse
|