Road Extraction Research Articles

Semisupervised Edge-Aware Road Extraction via Cross Teaching Between CNN and Transformer

Semisupervised Edge-Aware Road Extraction via Cross Teaching Between CNN and Transformer

Road Topology Extraction From Satellite Imagery by Joint Learning of Nodes and Their Connectivity

Road topology extraction from satellite images, which has long been of interest, is an essential task in remote sensing. The graph representation of road networks is one of the most challenging aspects of road topology extraction. Most existing approaches cast road extraction as binary segmentation and then use post-processing, such as skeletonization, to infer networks from pixel-wise prediction. In our work, we believe that a road network can be represented by an undirected graph denoted as G=(V,E), where V and E represent the set of road nodes and the set of edges between nodes, respectively. Thus, to construct the road topology, we propose NodeConnect, a new method of extracting nodes for a road network and inferring the connectivity between nodes. A convolutional neural network is jointly trained to predict the nodes and connectivity map for nodes, and the edges between nodes are inferred from the connectivity map. We compare our approach with several segmentation methods on the DeepGlobe and RoadTracer datasets. The experiments show that our approach achieves state-of-the-art performance in terms of pixel-based metrics and topological precision and recall.

A Road Extraction Method of a High-Resolution Remote Sensing Image Based on Multi-Feature Fusion and the Attention Mechanism

A Road Extraction Method of a High-Resolution Remote Sensing Image Based on Multi-Feature Fusion and the Attention Mechanism

Principles of processing and three-dimensional modelling through lidar data for applied research of the urban environment

Introduction. The 3D modeling technology of the urban environment using LiDAR survey data expands the possibilities of urban research. With proper use of various methods, models and algorithms for processing and analyzing LiDAR data, they can significantly facilitate and open up new opportunities for many applications discussed in this paper. The main research objective of the paper is to review methods for analyzing LiDAR survey data in urban studies and to present individual elements of the author’s optimization of these methods. Results. LiDAR data obtained as a result of laser scanning of the earth's surface from a certain vehicle form a three-dimensional terrain model in the point cloud form of varying density degrees. The post-processing of such data can branch out into many applications, which are discussed in this paper. The building extraction from a cloud of LiDAR points is performed using complex computational operations, the essence of which is to calculate the points of separate planes of the buildings roofs and then extract these points for 3D building modeling. There are many approaches to building extraction that aim to either improve the quality and accuracy of the extracted models or to speed up the data processing. Finding the optimal solution for 3D modeling of the urban environment is an urgent task in this area of research. Tracking changes in urban buildings involves comparing digital models of urban areas for different time periods in order to obtain the changes volume for each building. In a similar fashion, LiDAR data is used to assess damage to buildings by creating random points on the buildings walls and comparing their displacements before and after the damage. The population estimate using LiDAR data is based on a comparison of population data for census tracts with data on the number, area and volume of buildings in the same tracts obtained from processed LiDAR data. As a result, the expected population in each individual building can be calculated. Roads extraction from LiDAR data is performed by creating an image of the LiDAR laser pulse intensity and then comparing this image with a digital surface model. The article provides an example of a scheme for such road extraction. In addition, methods for extracting and mapping power lines by filtering the corresponding points are also considered. The ability to determine the exact size, slope, and exposure of a building's roof plane also makes it possible to estimate the potential level of solar radiation received by the roof, which can contribute to the optimal placement of solar power plants. Such an assessment may cause some difficulties, which are discussed in the article. The article proposes various optimization solutions for the considered methods, which were partially implemented in the ELiT software. In addition to effective tools for automatic data processing, the ELiT Project also provides an environment for high-quality visualization of results in a standard web-GIS interface. Conclusions. LiDAR data, in combination with efficient algorithms for processing and filtering data, greatly facilitates the solution of a number of tasks related to area monitoring and urban planning. In the future, the high accuracy of LiDAR data and the possibility of their visualization in GIS will make it possible to analyze the urban development features in order to identify the urban geosystemic properties of the city.

A new approach for road extraction using data augmentation and semantic segmentation

Accurate road extraction from remote sensing images is a challenging task. Several methods of extraction have been developed but the precision of extraction is still limited for the unpaved and small-width roads. This paper proposes an accurate road extraction approach called DAA-SSEG since it uses data augmentation architecture (DAA) and semantic segmentation model (SSEG). The proposed approach DAA-SSEG is based on a modified full convolutional neural network that overcomes the vanishing gradient and the training saturation issues. It recognizes roads at the pixel level. Furthermore, The DAA-SSEG approach uses a new plan of data augmentation based on geometric transformation and images refinement techniques. It allows getting a richer dataset thus better training and an accurate extraction. The experiment denotes that the proposed approach DAA-SSEG, that combine data augmentation architecture and semantic segmentation method, outperforms some state-of-the-art methods in terms of F-measures. The results demonstrate that it ensures accurate extraction of unpaved and small-width roads, in urban and rural areas. Moreover, the proposed approach distinguishes between roads and trails and can extract some roads not labeled beforehand.

Automated road extraction using reinforced road indices for Sentinel-2 data

Accurate Road feature extraction from multispectral images tends to be difficult due to the technical challenges posed by the data. The field of study has extreme importance with respect to urban planning, asset mapping, transport planning, and feasibility studies. In this paper, we utilize the road index (RI) to formulate our methodology for feature extraction. For further processing of the RI output image, we opted for Bot-Hat morphological operator to highlight the road features and suppress the surrounding features. We then classified the image using a Pixel-Based Unsupervised Classification to break the features into different classes. Subsequently, Contrast-Split segmentation algorithm was applied to segment the road network for network extraction. The experiment results showed an F1 score of 76.10% and 83.81% for the Lahore and Richmond datasets respectively.

Research on Road Extraction Method Based on Sustainable Development Goals Satellite-1 Nighttime Light Data

Road information plays a fundamental role in many applications. However, at present, it is difficult to extract road information from the traditional nighttime light images in view of their low spatial and spectral resolutions. To fill the gap in high-resolution nighttime light (NTL) data, the Sustainable Development Goals Satellite-1(SDGSAT-1) developed by the Chinese Academy of Sciences (CAS) was successfully launched on 5 November 2021. With 40 m spatial resolution, NTL data acquired by the Glimmer Imager Usual (GIU) sensor on the SDGSAT-1 provide a new data source for road extraction. To evaluate the ability of SDGSAT-1 NTL data to extract road information, we proposed a new road extraction method named Band Operation and Marker-based Watershed Segmentation Algorithm (BO-MWSA). Comparing with support vector machine (SVM) and optimum threshold (OT) algorithms, the results showed that: (1) the F1 scores of the roads in the test area extracted by SVM, OT, and BO-MWSA were all over 70%, indicating that SDGSAT-1/GIU data could be used as a data source for road extraction. (2) The F1 score of road extraction by BO-MWSA is 84.65%, which is 11.02% and 9.43% higher than those of SVM and OT, respectively. In addition, the F1 scores of BO-MWSA road extraction in Beijing and Wuhan are both more than 84%, indicating that BO-MWSA is an effective method for road extraction using NTL imagery. (3) In road extraction experiments for Lhasa, Beijing, and Wuhan, the results showed that the greater the traffic flow was, the lower the accuracy of the extracted roads became. Therefore, BO-MWSA is an effective method for road extraction using SDGSAT-1 NTL data.

GA-Net: A geometry prior assisted neural network for road extraction

With geospatial intelligence research developing rapidly, automatic road extraction is becoming a fundamental and challenging task. Due to the special geometric structure and spectral information of road networks, existing methods suffer from incomplete and fractured results. In this work, a novel road extraction convolutional neural network, incorporating the road boundary details and road junction information via a dual-branch multi-task structure, is proposed to learn synergistic feature representations and strengthen road connectivity. Firstly, a BiFPN-based feature aggregation module is utilised to bridge the semantic gap between low-level and high-level feature maps, allowing multi-scale spatial details to be fully fused. Secondly, the boundary auxiliary branch, using a U-shaped network with a spatial-channel attention module, captures residential information for the backbone to enhance the subtleties of road edges. Thirdly, the node inferring branch models the road junction position jointly with the road surface, aiming to strengthen the topology structure and reduce the fragmented road segments. We perform experiments on three diverse road datasets, namely the DeepGlobe dataset, Massachusetts dataset, and SpaceNet dataset. The results demonstrate that our model shows an overall performance improvement over some SOTA algorithms and the IoU indicator achieves 3.86%, 0.79%, and 1.71% improvements over Unet on the three datasets, respectively. • A novel Geometric Prior Assisted Neural Network (GA-Net) is proposed for automatic road extraction. The boundary information is introduced to the model, which strengthens the geometric details of road edges. The road junction information is jointly learned with the road surface information, which reduces the missing gap between road segments. • A feature fusion module is designed to bridge the semantic gap between multi-scale features maps. An attention module is designed to strengthen the road geometric information. The two light modules can be easily ensemble with other models. • We achieves better results than some SOTA algorithms for three popular datasets. And results of our model shows better completeness and connectivity.

Research on Road Extraction Algorithm Based on Improved Otsu’s Thresholding Method

In this paper, the widely used method of Otsu's thresholding method is studied. On the basis of the traditional Otsu's thresholding method, according to the distribution characteristics of the target and background in the actual image, a new method based on traditional Otsu's thresholding method is presented.and the road extraction experiment is carried out with GF-2 remote sensing image.The experimental results show that the improved method has the advantages of high segmentation accuracy and strong anti noise ability compared with the traditional Otsu's thresholding method.

Satellite IoT Based Road Extraction from VHR Images Through Superpixel-CNN Architecture

In the past few decades, technology has progressively become ineluctable in human lives, primarily due to the growth of certain fields like space technology, Big Data, the Internet of Things (IoT), and machine learning. Space technology has revolutionized communication mechanisms while creating opportunities for various research areas, including remote sensing (RS)-inspired applications. On the other hand, IoT presents a platform to use the power of the internet over a whole range of devices through a phenomenon known as social IoT. These devices generate a humongous amount of data that requires handling and managing by big data technology incorporated with deep learning techniques to reduce the manual workload of an operator. Moreover, deep learning architectures like convolutional neural networks (CNNs) have presented a scope to extract the underlying features from the large-scale input images in providing better solutions for tasks such as automatic road detection that come at the cost of time and memory overhead. In this context, we have proposed a three-layer edge-fog-cloud-based intelligent satellite IoT architecture that uses the superpixel-based CNN approach. At the fog layer, the superpixel-based simple linear iterative cluster (SLIC) algorithm uses the images captured by the satellites of the edge level to produce the smaller-sized superpixel images that can be transferred even in a low bandwidth link. The CNN module at the cloud level is then trained with these superpixel images to predict the road networks from these RS images. Two popular road datasets: the DeepGlobe Road dataset and the Massachusetts Road dataset, have been considered to prove the usefulness of the proposed SLIC-CNN architecture in satellite-based IoT platforms to address the problems like RS image-based road extraction. The proposed architecture achieves better performance accuracy than the classical CNN while reducing the incurred overhead by a noticeable limit. • Presents a superpixel-inspired deep learning-based architecture that optimizes the time and memory overhead. • Presents the three-layer edge-fog-cloud-based satellite IoT architecture to deploy the proposed SLIC-CNN framework. • Discusses the efficiency of the proposed deep architecture to work in a low bandwidth environment for road segmentation. • Highlights the application areas such as emergency alarm systems that can use the architecture to operate in real-time.

UGRoadUpd: An Unchanged-Guided Historical Road Database Updating Framework Based on Bi-Temporal Remote Sensing Images

Timely updated road networks are the basis for many real-world applications such as intelligent navigation and traffic management. Existing road updating methods based on remote sensing images learn from historical road databases to update roads. Road extraction models learned from historical images however, are not easily applied to a current image due to spectral differences; and only changed roads need updating. In this paper, an Unchanged-Guided Road Updating (UGRoadUpd) framework is proposed to improve the quality of updated road networks by limiting the road updating range and learning from historical unchanged roads. The UGRoadUpd framework identifies road changes using a novel dual-task dominant-transformer-based neural network for road change detection (DT-RoadCDNet). DT-RoadCDNet executes road segmentation and change detection simultaneously, from bi-temporal remote sensing images. The Dominant-Transformer based Global Context Modeling module in DT-RoadCDNet globally models the contextual spatial structure for improved integrity in roads and road changes. Based on the discovery of road changes, an unchanged-guided road update strategy updates the roads in changed areas by learning from the prior information provided by unchanged roads in a historical road database. Experiments on two newly annotated road change detection and update datasets confirms the effectiveness of our UGRoadUpd framework.

Corrigendum to “Exploring multiple crowdsourced data to learn deep convolutional neural networks for road extraction” [Int. J. Appl. Earth Observ. Geoinf. 104 (2021) 102544

Corrigendum to “Exploring multiple crowdsourced data to learn deep convolutional neural networks for road extraction” [Int. J. Appl. Earth Observ. Geoinf. 104 (2021) 102544

GMR-Net: Road-Extraction Network Based on Fusion of Local and Global Information

Road extraction from high-resolution remote-sensing images has high application values in various fields. However, such work is susceptible to the influence of the surrounding environment due to the diverse slenderness and complex connectivity of roads, leading to false judgment and omission during extraction. To solve this problem, a road-extraction network, the global attention multi-path dilated convolution gated refinement Network (GMR-Net), is proposed. The GMR-Net is facilitated by both local and global information. A residual module with an attention mechanism is first designed to obtain global and other aggregate information for each location’s features. Then, a multi-path dilated convolution (MDC) approach is used to extract road features at different scales, i.e., to achieve multi-scale road feature extraction. Finally, gated refinement units (GR) are proposed to filter out ambiguous features for the gradual refinement of details. Multiple road-extraction methods are compared in this study using the Deep-Globe and Massachusetts datasets. Experiments on these two datasets demonstrate that the proposed method achieves F1-scores of 87.38 and 85.70%, respectively, outperforming other approaches on segmentation accuracy and generalization ability.

ROAD EXTRACTION BASED ON IMPROVED DEEPLABV3 PLUS IN REMOTE SENSING IMAGE

Abstract. Urban roads in remote sensing images will be disturbed by surrounding ground features such as building shadows and tree shadows, and the extraction results are prone to problems such as incomplete road structure, poor topological connectivity, and poor accuracy. For mountain roads, there will also be problems such as hill shadow or vegetation occlusion. We propose an improved Deeplabv3+ semantic segmentation network method. This method uses ResNeSt, which introduces channel attention, as the backbone network, and combines the ASPP module to obtain multi-scale information, thereby improving the accuracy of road extraction. Analysis of the experimental results on the Deeplglobe dataset shows that the intersection ratio and accuracy of the method in this paper are 63.15% and 73.16%, respectively, which are better than other methods.

Graph-Based Image Segmentation for Road Extraction from Post-Disaster Aerial Footage

This research effort proposes a novel method for identifying and extracting roads from aerial images taken after a disaster using graph-based image segmentation. The dataset that is used consists of images taken by an Unmanned Aerial Vehicle (UAV) at the University of West Florida (UWF) after hurricane Sally. Ground truth masks were created for these images, which divide the image pixels into three categories: road, non-road, and uncertain. A specific pre-processing step was implemented, which used Catmull–Rom cubic interpolation to resize the image. Moreover, the Gaussian filter used in Efficient Graph-Based Image Segmentation is replaced with a median filter, and the color space is converted from RGB to HSV. The Efficient Graph-Based Image Segmentation is further modified by (i) changing the Moore pixel neighborhood to the Von Neumann pixel neighborhood, (ii) introducing a new adaptive isoperimetric quotient threshold function, (iii) changing the distance function used to create the graph edges, and (iv) changing the sorting algorithm so that the algorithm can run more effectively. Finally, a simple function to automatically compute the k (scale) parameter is added. A new post-processing heuristic is proposed for road extraction, and the Intersection over Union evaluation metric is used to quantify the road extraction performance. The proposed method maintains high performance on all of the images in the dataset and achieves an Intersection over Union (IoU) score, which is significantly higher than the score of a similar road extraction technique using K-means clustering.

Image-to-Image Subpixel Registration Based on Template Matching of Road Network Extracted by Deep Learning

The vast digital archives collected by optical remote sensing observations over a long period of time can be used to determine changes in the land surface and this information can be very useful in a variety of applications. However, accurate change extraction requires highly accurate image-to-image registration, which is especially true when the target is urban areas in high-resolution remote sensing images. In this paper, we propose a new method for automatic registration between images that can be applied to noisy images such as old aerial photographs taken with analog film, in the case where changes in man-made objects such as buildings in urban areas are extracted from multitemporal high-resolution remote sensing images. The proposed method performs image-to-image registration by applying template matching to road masks extracted from images using a two-step deep learning model. We applied the proposed method to multitemporal images, including images taken more than 36 years before the reference image. As a result, the proposed method achieved registration accuracy at the subpixel level, which was more accurate than the conventional area-based and feature-based methods, even for image pairs with the most distant acquisition times. The proposed method is expected to provide more robust image-to-image registration for differences in sensor characteristics, acquisition time, resolution and color tone of two remote sensing images, as well as to temporal variations in vegetation and the effects of building shadows. These results were obtained with a road extraction model trained on images from a single area, single time period and single platform, demonstrating the high versatility of the model. Furthermore, the performance is expected to be improved and stabilized by using images from different areas, time periods and platforms for training.

Road Extraction Based on Improved Convolutional Neural Networks with Satellite Images

Deep learning has been applied in various fields for its effective and accurate feature learning capabilities in recent years. Currently, information extracted from remote sensing images with the learning methods has become the most relevant research area for its developed precision. In terms of developing segmentation precision and reducing calculation power consumption, the improved deep learning methods have received more attention, and the improvement of semantic segmentation architectures has been a popular solution. This research presents a learning method named D-DenseNet with a new structure for road extraction. The methods for the improvement are divided into two stages: (1) alternate the consecutive dilated convolutions number in the structure of the network (2) the stem block is arranged as the initial block. So, dilated convolution can obtain more global context information through the whole network. Further, the D-DenseNet restructures D-LinkNet by taking DenseNet as its backbone instead of ResNet, which can expand the receptive field and accept more feature information. The D-DenseNet is effective because of its 119 M model size and 57.96% IoU on the processing test data and 99.3 M modes size and 66.26% on the public dataset, which achieved the research objective for reducing model size and developing segmentation precision—IoU. The experiment indicates that the D-Dense block and the stem block are effective for developing road extraction, and the appropriate number of convolution layers is also essential for model evaluation.

MECA-Net: A MultiScale Feature Encoding and Long-Range Context-Aware Network for Road Extraction from Remote Sensing Images

Road extraction from remote sensing images is significant for urban planning, intelligent transportation, and vehicle navigation. However, it is challenging to automatically extract roads from remote sensing images because the scale difference of roads in remote sensing images varies greatly, and slender roads are difficult to identify. Moreover, the road in the image is often blocked by the shadows of trees and buildings, which results in discontinuous and incomplete extraction results. To solve the above problems, this paper proposes a multiscale feature encoding and long-range context-aware network (MECA-Net) for road extraction. MECA-Net adopts an encoder–decoder structure and contains two core modules. One is the multiscale feature encoding module, which aggregates multiscale road features to improve the recognition ability of slender roads. The other is the long-range context-aware module, which consists of the channel attention module and the strip pooling module, and is used to obtain sufficient long-range context information from the channel dimension and spatial dimension to alleviate road occlusion. Experimental results on the open DeepGlobe road dataset and Massachusetts road dataset indicate that the proposed MECA-Net outperforms the other eight mainstream networks, which verifies the effectiveness of the proposed method.

Road extraction through digital processing and visual interpretation of satellite images

Road extraction plays an important role in urban planning and city extension issues, as well as in road monitoring, traffic management and map updating. Technological advances may offer a wealth of data and techniques that could be implemented in road delineation and extraction, as well as in change detection projects. Among the various methods that have been developed for this purpose, remote sensing techniques and especially digital processing of satellite data could contribute significantly in this direction. This paper presents the study of a road network which concerns the city of Kastoria located in northwestern Greece and its surrounding area. The special character of the road network is closely related to the city’s cultural value which is depicted in its structure. Satellite images of various spatial resolutions were employed for digital processing. Landsat 8 imagery was used in order to detect and delineate the linear features through spatial enhancement, while Landsat 5 images were used to detect changes over time through visual interpretation. Semi-automatic techniques were applied to SPOTmaps products to extract the road network, while an object-oriented approach was applied to QuickBird imagery in order to combine the spatial components with spectral properties. Through semi-automatic digitization and object-oriented workflows, the export of the studied part of the road network in vector format was achieved, thus facilitating the process and reducing the required time. The resulted data are efficient in road network delineation and could be combined with other data for road maintenance and extension, change detection issues, as well as for cultural and touristic purposes.

A Land Cover Background-Adaptive Framework for Large-Scale Road Extraction

Background: Road network data are crucial in various applications, such as emergency response, urban planning, and transportation management. The recent application of deep neural networks has significantly boosted the efficiency and accuracy of road network extraction based on remote sensing data. However, most existing methods for road extraction were designed at local or regional scales. Automatic extraction of large-scale road datasets from satellite images remains challenging due to the complex background around the roads, especially the complicated land cover types. To tackle this issue, this paper proposes a land cover background-adaptive framework for large-scale road extraction. Method: A large number of sample image blocks (6820) are selected from six different countries of a wide region as the dataset. OpenStreetMap (OSM) is automatically converted to the ground truth of networks, and Esri 2020 Land Cover Dataset is taken as the background land cover information. A fuzzy C-means clustering algorithm is first applied to cluster the sample images according to the proportion of certain land use types that obviously negatively affect road extraction performance. Then, the specific model is trained on the images clustered as abundant with that certain land use type, while a general model is trained based on the rest of the images. Finally, the road extraction results obtained by those general and specific modes are combined. Results: The dataset selection and algorithm implementation were conducted on the cloud-based geoinformation platform Google Earth Engine (GEE) and Google Colaboratory. Experimental results showed that the proposed framework achieved stronger adaptivity on large-scale road extraction in both visual and statistical analysis. The C-means clustering algorithm applied in this study outperformed other hard clustering algorithms. Significance: The promising potential of the proposed background-adaptive network was demonstrated in the automatic extraction of large-scale road networks from satellite images as well as other object detection tasks. This search demonstrated a new paradigm for the study of large-scale remote sensing applications based on deep neural networks.