Road Extraction Method Research Articles

An Ensemble Wasserstein Generative Adversarial Network Method for Road Extraction From High Resolution Remote Sensing Images in Rural Areas

Road extraction from high resolution remote sensing (HR-RS) images is an important yet challenging computer vision task. In this study, we propose an ensemble Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) method called E-WGAN-GP for road extraction from HR-RS images in rural areas. The WGAN-GP model modifies the standard GANs with Wasserstein distance and gradient penalty. We add a spatial penalty term in the loss function of the WGAN-GP model to solve the class imbalance problem typically in road extraction. Parameter experiments are undertaken to determine the best spatial penalty and the weight term in the new loss function based on GaoFen-2 dataset. In addition, we execute an ensemble strategy in which we first train two WGAN-GP models using the U-Net and BiSeNet as generator respectively, and then intersect their inferred outputs to yield better road vectors. We train our new model with GaoFen-2 HR-RS images in rural areas from China and also the DeepGlobe Road Extraction dataset. Compared with the U-Net, BiSeNet, D-LinkNet and WGAN-GP methods without ensemble, our new method makes a good trade-off between precision and recall with F1-score = 0.85 and IoU = 0.73.

USING OF VHR SATELLITE IMAGES FOR ROAD NETWORK EXTRACTION IN EGYPT

Roads extraction from VHR satellite images are very paramount for GIS and map updating. Due to the high resolution of satellite images, there are many obstacles broken roads such as shadow, and vehicles. The present work aims to find the most suitable road extraction approach that can be applied in the Egyptian environment. In this study, two satellite images from WorldView-2 and WorldView-3 were used. Classification of image by pixel-based and object-based was carried out to find the appropriate classification method for road extraction. Then, road class refinement by morphology and angular texture signature are performed to decrease the misclassifications between roads and other spectrally similar objects. After that, an iterative and localized Hough transform method was compared with the thinning algorithm method to find the proper method that can extract road centerline segments from the refined images. The performance of the extracted roads was estimated by using the common metrics; completeness, correctness, and quality. The results of this work demonstrate that the random tree in object-based classification achieves the highest overall accuracy than other classification methods. Also, thinning algorithm has more advantages than Hough transform.

A Self-Supervised Learning Framework for Road Centerline Extraction From High-Resolution Remote Sensing Images

Road extraction from the high-resolution remote sensing image is significant for the land planning, vehicle navigation, etc. The existing road extraction methods normally need many preprocessing and subsequent optimization steps. Therefore, an automatic road centerline extraction method based on the self-supervised learning framework for high-resolution remote sensing image is proposed. This proposed method does not need to manually select training samples and other optimization steps, such as the nonroad area removing. First, the positive sample selection method combining the spectral and shape features is proposed to extract the road sample. Then, the one-class classifier framework is introduced and the random forest positive unlabeled learning classifier is constructed to get the posterior probability of the pixel belonging to road. The shape feature and the posterior probability are combined to form the final road network in the object-oriented way. Finally, the road centerline is obtained through the tensor voting algorithm. In order to verify the effectiveness of the proposed algorithm, high-resolution remote sensing images and benchmark datasets are used to do experiments. The indexes of the completeness ratio, the correctness ratio, and the detection quality are used for the quantitative accuracy evaluation. Compared with the supervised, the unsupervised, and the one-class classification road extraction algorithms, this proposed algorithm achieves high accuracy and efficiency. For the deep learning method comparison, the deep learning method performs well in most cases especially in the complex urban area. However, the deep learning method needs a large number of samples and a long training time, and our self-supervised learning framework does not need the training samples.

Development and prospect of road extraction method for optical remote sensing image

Development and prospect of road extraction method for optical remote sensing image

Road Extraction From High-Resolution Satellite Images Based on Multiple Descriptors

Geometry and texture noise make it difficult to accurately describe road image rules, which leads to the low degree of automation of traditional template matching algorithms based on internal texture homogenization. We propose a semi-automatic road extraction method based on multiple descriptors to improve the degree of automation while ensuring the accuracy of road extraction. This method aims to address the problems of incomplete road image geometric information and poor homogeneity of internal road texture. The multiscale line segment orientation histogram model and sector descriptor are established. Road points are tracked by interpolation and extension, and postprocessing is used to fit the tracking points and extract the road routes. In this article, high-resolution remote sensing images of different types, different resolutions, and different scenes are selected, and the roads exhibit curvatures, vehicle and shadow occlusions, roundabouts, and variational features. Experiments show that for roads with a certain width, completeness, and correctness of the method are more than 98%. Additionally, as compared with other algorithms, the interactive human intervention of this method is reduced by more than 2/3.

结合卷积神经网络和张量投票的道路提取方法

结合卷积神经网络和张量投票的道路提取方法

Fully Convolutional Network-Based Ensemble Method for Road Extraction From Aerial Images

This letter proposed a road extraction method based on fully convolutional networks (FCNs) with an ensemble strategy in order to solve the imbalance of road and background areas in aerial images. By utilizing the FCN, we consider road extraction as a semantic segmentation problem. In the network, the weight of the loss function is modified because of the imbalance between the roads and backgrounds, and there will be a larger punishment if roads are wrongly classified as background. Since it is difficult to determine an appropriate weight of the loss function for a given image, an ensemble method based on spatial consistency (SC) is proposed. The result maps that are obtained from the FCNs with different loss functions are fused in our proposed ensemble strategy, which also avoids the determination of weights. Our method is tested using the Massachusetts road data set, and it was proven to be effective compared with the base fully convolutional model according to our experimental result.

HsgNet: A Road Extraction Network Based on Global Perception of High-Order Spatial Information

Road extraction is a unique and difficult problem in the field of semantic segmentation because roads have attributes such as slenderness, long span, complexity, and topological connectivity, etc. Therefore, we propose a novel road extraction network, abbreviated HsgNet, based on high-order spatial information global perception network using bilinear pooling. HsgNet, taking the efficient LinkNet as its basic architecture, embeds a Middle Block between the Encoder and Decoder. The Middle Block learns to preserve global-context semantic information, long-distance spatial information and relationships, and different feature channels’ information and dependencies. It is different from other road segmentation methods which lose spatial information, such as those using dilated convolution and multiscale feature fusion to record local-context semantic information. The Middle Block consists of three important steps: (1) forming a feature resource pool to gather high-order global spatial information; (2) selecting a feature weight distribution, enabling each pixel position to obtain complementary features according to its own needs; and (3) inversely mapping the intermediate output feature encoding to the size of the input image by expanding the number of channels of the intermediate output feature. We compared multiple road extraction methods on two open datasets, SpaceNet and DeepGlobe. The results show that compared to the efficient road extraction model D-LinkNet, our model has fewer parameters and better performance: we achieved higher mean intersection over union (71.1%), and the model parameters were reduced in number by about 1/4.

A Method for Road Extraction from High-Resolution Remote Sensing Images Based on Multi-Kernel Learning

Extracting road from high resolution remote sensing (HRRS) images is an economic and effective way to acquire road information, which has become an important research topic and has a wide range of applications. In this paper, we present a novel method for road extraction from HRRS images. Multi-kernel learning is first utilized to integrate the spectral, texture, and linear features of images to classify the images into road and non-road groups. A precise extraction method for road elements is then designed by building road shaped indexes to automatically filter out the interference of non-road noises. A series of morphological operations are also carried out to smooth and repair the structure and shape of the road element. Finally, based on the prior knowledge and topological features of the road, a set of penalty factors and a penalty function are constructed to connect road elements to form a complete road network. Experiments are carried out with different sensors, different resolutions, and different scenes to verify the theoretical analysis. Quantitative results prove that the proposed method can optimize the weights of different features, eliminate non-road noises, effectively group road elements, and greatly improve the accuracy of road recognition.

Automated Method of Extracting Urban Roads Based on Region Growing from Mobile Laser Scanning Data.

With the rapid development of three-dimensional point cloud acquisition from mobile laser scanning systems, the extraction of urban roads has become a major research focus. Although it has great potential for digital image processing, the extraction of roads using the region growing approach is still in its infancy. We propose an automated method of urban road extraction based on region growing. First, an initial seed is chosen under constraints relating to the Gaussian curvature, height and number of neighboring points, which ensures that the initial seed is located on a road. Then, the growing condition is determined by the angle threshold of the tangent plane of the seed point. Then, new seeds are selected based on the identified road points and their curvature. The method also includes a strategy for dealing with multiple discontinuous roads in a dataset. The result shows that the method can not only achieve high accuracy in urban road extraction but is also stable and robust.

A Method for Road Network Extraction from High-Resolution SAR Imagery Using Direction Grouping and Curve Fitting

Roads are an important recognition target in synthetic aperture radar (SAR) image interpretation. Although a considerable number of high-quality SAR images are now available, the method of road extraction is lagging. To extract the road network with low missed and false rates, this paper proposed a road network extraction approach which includes line detection, road segmentation, road network extraction and optimization. First, the linear feature response and direction map are obtained from the SAR intensity image using the multiplicative Duda operation. Then, the backscattering coefficient and coefficient of variation are combined using a support vector machine to eliminate the linear structures of non-roads, and the binary image of road candidates is subsequently achieved by morphological profiles of path openings. Next, with the obtained direction map, a novel thinning method based on binary image decomposition and curve fitting is presented to obtain line segments of the road network. Finally, a series of measures which involve overlap, continuity, and junction optimization are proposed to construct the road network. In the experiments, the proposed method was applied to Radarsat-2 and TerraSAR-X high-resolution images. The experimental results showed that the proposed method had an excellent performance in terms of both completeness and correctness.

Lane-Level Road Extraction from High-Resolution Optical Satellite Images

High-quality updates of road information play an important role in smart city planning, sustainable urban expansion, vehicle management, urban planning, traffic navigation, public health and other fields. However, due to interference from road geometry and texture noise, it is difficult to avoid the decline of automation while accurately extracting roads. Therefore, we propose a high-resolution optical satellite image lane-level road extraction method. First, from the perspective of template matching and considering road characteristics and relevant semantic relations, an adaptive correction model, an MLSOH (multi-scale line segment orientation histogram) descriptor, a sector descriptor, and a multiangle beamlet descriptor are proposed to solve the interference from geometry and texture noise in road template matching and tracking. Second, based on refined lane-level tracking, single-lane and double-lane road-tracking modes are designed to extract single-lane and double-lane roads, respectively. In this paper, Pleiades satellite and GF-2 images are selected to set up different scenarios for urban and rural areas. Experiments are carried out on the phenomena that restrict road extraction, such as tree occlusion, building shadow occlusion, road bending, and road boundary blurring. Compared with other methods, the proposed method not only ensures the accuracy of lane-level road extraction but also greatly improves the automation of road extraction.

Automated Method of Road Extraction from Aerial Images Using a Deep Convolutional Neural Network

Updating road networks using remote sensing imagery is among the most important topics in city planning, traffic management and disaster management. As a good alternative to manual methods, which are considered to be expensive and time consuming, deep learning techniques provide great improvements in these regards. One of these techniques is the use of deep convolution neural networks (DCNNs). This study presents a road segmentation model consisting of a skip connection of U-net and residual blocks (ResBlocks) in the encoding part and convolution layers (Conv. layer) in the decoding part. Although the model uses fewer residual blocks in the encoding part and fewer convolution layers in the decoding part, it produces better image predictions in comparison with other state-of-the-art models. This model automatically and efficiently extracts road networks from high-resolution aerial imagery in an unexpansive manner using a small training dataset.

Multi-Scale and Multi-Task Deep Learning Framework for Automatic Road Extraction

Road detection and centerline extraction from very high-resolution (VHR) remote sensing imagery are of great significance in various practical applications. Road detection and centerline extraction operations depend on each other, to a certain extent. The road detection constrains the appearance of the centerline, and the centerline enhances the linear features of the road detection. However, most of the previous works have addressed these two tasks separately and have not considered the symbiotic relationship between them, making it difficult to obtain smooth and complete roads. In this paper, a novel multi-scale and multi-task deep learning framework for automatic road extraction (MSMT-RE) is proposed to build the relationship between them and simultaneously complete the road detection and centerline extraction tasks. U-Net is selected as the basic network for multi-task learning due to its strong ability to preserve spatial details. Multi-scale feature integration is also applied in the framework to increase the robustness of the feature extraction. Meanwhile, an adaptive loss function is introduced to solve the problems of roads taking up a small percentage of the training samples, and the fact that the positive samples of the two tasks are unbalanced. Finally, experiments were conducted on two public road data sets and two large images from Google Earth, and the proposed framework was compared with other state-of-the-art deep learning-based road extraction methods, both quantitatively and qualitatively. The proposed approach outperformed all the compared methods, confirming its advantages in automatic road extraction.

A Multiple Feature Fully Convolutional Network for Road Extraction From High-Resolution Remote Sensing Image Over Mountainous Areas

Road extraction from the remote sensing image over mountainous areas is a difficult vision problem. In this letter, we propose a multiple feature fully convolutional network (MFFCN) on the basis of FCN for mountainous road extraction. The benefits of this model are twofold: first, MFFCN is a semantic segmentation model, which has deep convolutional networks. It avoids the problem of repeated storage and computational convolutions caused by the use of pixel blocks. Second, the MFFCN model could extract the spectral and terrain features. This method ensures the integrity and continuity of the road extraction results. The dataset is composed of GF-2 data and ASTER GDEM data in the Shigatse region of Tibet. We test our network on the dataset and compare it with four road extraction methods. The result shows that the proposed MFFCN is superior to all the comparing methods.

Extracting Flooded Roads by Fusing GPS Trajectories and Road Network

Urban roads are the lifeline of urban transportation and satisfy the commuting and travel needs of citizens. Following the acceleration of urbanization and the frequent extreme weather in recent years, urban waterlogging is occurring more than usual in summer and has negative effects on the urban traffic networks. Extracting flooded roads is a critical procedure for improving the resistance ability of roads after urban waterlogging occurs. This paper proposes a flooded road extraction method to extract the flooding degree and the time at which roads become flooded in large urban areas by using global positioning system (GPS) trajectory points with driving status information and the high position accuracy of vector road data with semantic information. This method uses partition statistics to create density grids (grid layer) and uses map matching to construct a time-series of GPS trajectory point density for each road (vector layer). Finally, the fusion of grids and vector layers obtains a more accurate result. The experiment uses a dataset of GPS trajectory points and vector road data in the Wuchang district, which proves that the extraction result has a high similarity with respect to the flooded roads reported in the news. Additionally, extracted flooded roads that were not reported in the news were also found. Compared with the traditional methods for extracting flooded roads and areas, such as rainfall simulation and SAR image-based classification in urban areas, the proposed method discovers hidden flooding information from geospatial big data, uploaded at no cost by urban taxis and remaining stable for a long period of time.

Aerial Image Road Extraction Based on an Improved Generative Adversarial Network

Aerial photographs and satellite images are one of the resources used for earth observation. In practice, automated detection of roads on aerial images is of significant values for the application such as car navigation, law enforcement, and fire services. In this paper, we present a novel road extraction method from aerial images based on an improved generative adversarial network, which is an end-to-end framework only requiring a few samples for training. Experimental results on the Massachusetts Roads Dataset show that the proposed method provides better performance than several state of the art techniques in terms of detection accuracy, recall, precision and F1-score.

An improved road extraction method for remote sensing images based on Canny edge detection

An improved road extraction method for remote sensing images based on Canny edge detection

A Novel Method of Missing Road Generation in City Blocks Based on Big Mobile Navigation Trajectory Data

With the rapid development of cities, the geographic information of urban blocks is also changing rapidly. However, traditional methods of updating road data cannot keep up with this development because they require a high level of professional expertise for operation and are very time-consuming. In this paper, we develop a novel method for extracting missing roadways by reconstructing the topology of the roads from big mobile navigation trajectory data. The three main steps include filtering of original navigation trajectory data, extracting the road centerline from navigation points, and establishing the topology of existing roads. First, data from pedestrians and drivers on existing roads were deleted from the raw data. Second, the centerlines of city block roads were extracted using the RSC (ring-stepping clustering) method proposed herein. Finally, the topologies of missing roads and the connections between missing and existing roads were built. A complex urban block with an area of 5.76 square kilometers was selected as the case study area. The validity of the proposed method was verified using a dataset consisting of five days of mobile navigation trajectory data. The experimental results showed that the average absolute error of the length of the generated centerlines was 1.84 m. Comparative analysis with other existing road extraction methods showed that the F-score performance of the proposed method was much better than previous methods.

Road Topology Refinement via a Multi-Conditional Generative Adversarial Network.

With the rapid development of intelligent transportation, there comes huge demands for high-precision road network maps. However, due to the complex road spectral performance, it is very challenging to extract road networks with complete topologies. Based on the topological networks produced by previous road extraction methods, in this paper, we propose a Multi-conditional Generative Adversarial Network (McGAN) to obtain complete road networks by refining the imperfect road topology. The proposed McGAN, which is composed of two discriminators and a generator, takes both original remote sensing image and the initial road network produced by existing road extraction methods as input. The first discriminator employs the original spectral information to instruct the reconstruction, and the other discriminator aims to refine the road network topology. Such a structure makes the generator capable of receiving both spectral and topological information of the road region, thus producing more complete road networks compared with the initial road network. Three different datasets were used to compare McGan with several recent approaches, which showed that the proposed method significantly improved the precision and recall of the road networks, and also worked well for those road regions where previous methods could hardly obtain complete structures.