Road Extraction Research Articles

AN INFORMAL ROAD DETECTION NEURAL NETWORK FOR SOCIETAL IMPACT IN DEVELOPING COUNTRIES

Abstract. Roads found in informal settlements arise out of convenience, and are often not recorded or maintained by authorities. This complicates service delivery, sustainable development and crisis mitigation, including management and tracking of COVID-19. We, therefore, aim to extract informal roads in remote sensing images. Existing techniques aiming at the extraction of formal roads are not suitable for the problem due to the complex physical and spectral properties of informal roads. The only existing approaches for informal roads, namely (Nobrega et al., 2006, Thiede et al., 2020), do not consider neural networks as a solution. Neural networks show promise in overcoming these complexities. However, they require a large amount of data to learn, which is currently not available due to the expensive and time-consuming nature of collecting such data. This paper implements a neural network to extract informal roads from a data set digitised by this research group. Data quality is assessed by calculating validity completeness, homogeneity and the V-measure, a measure of consistency, in order to evaluate the overall usability of the dataset for neural network informal road detection. We implement the GANs-UNet model that obtained the highest F1-score in a 2020 review paper (Abdollahi et al., 2020) on the state-of-the-art deep learning models used to extract formal roads. The results indicate that the model is able to extract informal roads successfully in the presence of appropriate training data.

A NOVEL DATA AUGMENTATION METHOD TO ENHANCE THE TRAINING DATASET FOR ROAD EXTRACTION FROM SWISS HISTORICAL MAPS

Abstract. Long-term retrospective road data are required for various analyses (e.g., investigation of urban sprawl, analysis of road network evolution). Yet, it is challenging to extract roads from scanned historical maps due to their dissatisfying quality. Although deep learning has been exerting its superiority in image segmentation, its application to road extraction from historical maps is rarely seen in existing studies. Deep learning usually requires quite large amounts of training data, which is time-consuming and tedious to label. Data augmentation can to some extent solve this issue. The existing data augmentation techniques vary each training sample as a whole (e.g., rotation, flipping). But some features or symbols on maps will never occur in practice when they are rotated or flipped (e.g., numbers, labels). To solve this problem and to further improve the diversity of training samples, we propose a novel data augmentation method, which varies the target features instead of the whole training sample. The method is validated by applying it to road extraction from the historical Swiss Siegfried map. The experiment results show the effectiveness of the proposed method.

Road Extraction Convolutional Neural Network with Embedded Attention Mechanism for Remote Sensing Imagery

Roads are closely related to people’s lives, and road network extraction has become one of the most important remote sensing tasks. This study aimed to propose a road extraction network with an embedded attention mechanism to solve the problem of automatic extraction of road networks from a large number of remote sensing images. Channel attention mechanism and spatial attention mechanism were introduced to enhance the use of spectral information and spatial information based on the U-Net framework. Moreover, residual densely connected blocks were introduced to enhance feature reuse and information flow transfer, and a residual dilated convolution module was introduced to extract road network information at different scales. The experimental results showed that the method proposed in this study outperformed the compared algorithms in overall accuracy. This method had fewer false detections, and the extracted roads were closer to ground truth. Ablation experiments showed that the proposed modules could effectively improve road extraction accuracy.

Extraction and Calculation of Roadway Area from Satellite Images Using Improved Deep Learning Model and Post-Processing.

Roadway area calculation is a novel problem in remote sensing and urban planning. This paper models this problem as a two-step problem, roadway extraction, and area calculation. Roadway extraction from satellite images is a problem that has been tackled many times before. This paper proposes a method using pixel resolution to calculate the area of the roads covered in satellite images. The proposed approach uses novel U-net and Resnet architectures called U-net++ and ResNeXt. The state-of-the-art model is combined with the proposed efficient post-processing approach to improve the overlap with ground truth labels. The performance of the proposed road extraction algorithm is evaluated on the Massachusetts dataset and it is shown that the proposed approach outperforms the existing solutions which use models from the U-net family.

MSPNet: Multi-Scale Strip Pooling Network for Road Extraction from Remote Sensing Images

Extracting roads from remote sensing images can support a range of geo-information applications. However, it is challenging due to factors such as the complex distribution of ground objects and occlusion of buildings, trees, shadows, etc. Pixel-wise classification often fails to predict road connectivity and thus produces fragmented road segments. In this paper, we propose a multi-scale strip pooling network (MSPNet) to learn the linear features of roads. Motivated by the strip pooling being more aligned with the shape of roads, which are long-span and narrow, we develop a multi-scale strip pooling (MSP) module that utilizes strip pooling layers with long but narrow kernel shapes to capture multi-scale long-range context from horizontal and vertical directions. The proposed MSP module focuses on establishing relationships along the road region to guarantee the connectivity of roads. Considering the complex distribution of ground objects, the spatial pyramid pooling is applied to enhance the learning ability of complex features in different sub-regions. In addition, to alleviate the problem caused by an imbalanced distribution of road and non-road pixels, we use binary cross-entropy and dice-coefficient loss functions to jointly train our proposed deep learning model. Then, we perform ablation experiments to adjust the loss contributions to suit the task of road extraction. Comparative experiments on a popular benchmark DeepGlobe dataset demonstrate that our proposed MSPNet establishes new competitive results in both IoU and F1-score.

Field road segmentation network based on PraNet

ABSTRACT Nowdays, many methods based on CNN have been proposed for road extraction. However, there are still great challenges. Therefore, according to image characteristics, this paper made corresponding improvements based on medical segmentation network PraNet. First, the reverse attention module (RA) connected at the last layer of PraNet is changed to the positive attention module (PA). Then, the negative matrix L1 norm regularization is added into the loss function. We conducted experiments on a data set made by UAV in the field of Henan Academy of Agricultural Sciences. The results show that the proposed method is better than the comparison models.

Bi-HRNet: A Road Extraction Framework from Satellite Imagery Based on Node Heatmap and Bidirectional Connectivity

Today, with the rapid development of the geographic information industry, automatic road extraction from satellite imagery is a basic requirement. Most existing methods have been designed based on binary segmentation. However, these methods do not consider the topological features of road networks, which include point, edge, and direction. In this study, a topology-based multi-task convolution network is designed, namely Bi-HRNet, which can effectively learn the key features of nodes and their directions. First, the proposed network learns the node heatmap of roads, and then the pixel coordinates are extracted from the node heatmap via non-maximum suppression (NMS). At the same time, the connectivity between nodes is predicted. To improve the integrity and accuracy of connectivity, we propose a bidirectional connectivity prediction strategy, which can learn the bidirectional categories instead of direction angles. The bidirectional categories are designed based on “top-to-bottom” and “bottom-to-top” strategies, which can improve the accuracy of the connectivity between nodes. To illustrate the effectiveness of the proposed Bi-HRNet, we compare our method with several methods on different datasets. The experiments show that our method achieves a state-of-the-art performance and significantly outperforms various previous methods.

Forest road extraction from orthophoto images by convolutional neural networks

Continuous monitoring of the forest road infrastructure and keeping track of the changes occurred are important for forestry practices, map updating, forest fire and forest transport decision support systems. In this context, the most of up to date data can be obtained by automatic forest road extraction from satellite images via machine learning (ML). Acquiring sufficient data is one of the most important factors which affect the success of ML and deep learning (DL). DL architectures yield more consistent results for complex data sets compared with ML algorithms. In the present study, three different deep learning (Resnet-18, MobileNet-V2 and Xception) architectures with semantic segmentation architecture were compared for extracting the forest road network from high-resolution orthophoto images and the results were analyzed. The architectures were evaluated through a multiclass statistical analysis based precision, recall, F1 score, intersection over union and overall accuracy (OA). The results present significant values obtained by the Resnet-18 architecture, with 99.72% of OA and 98.87% of precision and by the MobileNet-V2 architecture, with 97.76% of OA and 98.28% of precision. Also the results show that Resnet-18, MobileNet-V2 semantic segmentation architectures can be used efficiently for forest road extraction.

Multi-Resolution Transformer Network for Building and Road Segmentation of Remote Sensing Image

Extracting buildings and roads from remote sensing images is very important in the area of land cover monitoring, which is of great help to urban planning. Currently, a deep learning method is used by the majority of building and road extraction algorithms. However, for existing semantic segmentation, it has a limitation on the receptive field of high-resolution remote sensing images, which means that it can not show the long-distance scene well during pixel classification, and the image features is compressed during down-sampling, meaning that the detailed information is lost. In order to address these issues, Hybrid Multi-resolution and Transformer semantic extraction Network (HMRT) is proposed in this paper, by which a global receptive field for each pixel can be provided, a small receptive field of convolutional neural networks (CNN) can be overcome, and the ability of scene understanding can be enhanced well. Firstly, we blend the features by branches of different resolutions to keep the high-resolution and multi-resolution during down-sampling and fully retain feature information. Secondly, we introduce the Transformer sequence feature extraction network and use encoding and decoding to realize that each pixel has the global receptive field. The recall, F1, OA and MIoU of HMPR obtain 85.32%, 84.88%, 85.99% and 74.19%, respectively, in the main experiment and reach 91.29%, 90.41%, 91.32% and 84.00%, respectively, in the generalization experiment, which prove that the method proposed is better than existing methods.

Survey of Road Extraction Methods in Remote Sensing Images Based on Deep Learning

Survey of Road Extraction Methods in Remote Sensing Images Based on Deep Learning

Mixer U-Net: An Improved Automatic Road Extraction from UAV Imagery

Automatic road extraction from unmanned aerial vehicle (UAV) imagery has been one of the major research topics in the area of remote sensing analysis due to its importance in a wide range of applications such as urban planning, road monitoring, intelligent transportation systems, and automatic road navigation. Thanks to the recent advances in Deep Learning (DL), the tedious manual segmentation of roads can be automated. However, the majority of these models are computationally heavy and, thus, are not suitable for UAV remote-sensing tasks with limited resources. To alleviate this bottleneck, we propose two lightweight models based on depthwise separable convolutions and ConvMixer inception block. Both models take the advantage of computational efficiency of depthwise separable convolutions and multi-scale processing of inception module and combine them in an encoder–decoder architecture of U-Net. Specifically, we substitute standard convolution layers used in U-Net for ConvMixer layers. Furthermore, in order to learn images on different scales, we apply ConvMixer layer into Inception module. Finally, we incorporate pathway networks along the skip connections to minimize the semantic gap between encoder and decoder. In order to validate the performance and effectiveness of the models, we adopt Massachusetts roads dataset. One incarnation of our models is able to beat the U-Net’s performance with 10× fewer parameters, and DeepLabV3’s performance with 12× fewer parameters in terms of mean intersection over union (mIoU) metric. For further validation, we have compared our models against four baselines in total and used additional metrics such as precision (P), recall (R), and F1 score.

SDUNet: Road extraction via spatial enhanced and densely connected UNet

Extracting road maps from high-resolution optical remote sensing images has received much attention recently, especially with the rapid development of deep learning methods. However, most of these CNN based approaches simply focused on multi-scale encoder architectures or multiple branches in neural networks, and ignored some inherent characteristics of the road surface. In this paper, we design a novel network for road extraction based on spatial enhanced and densely connected UNet, called SDUNet. SDUNet aggregates both the multi-level features and global prior information of road networks by combining the strengths of spatial CNN-based segmentation and densely connected blocks. To enhance the feature learning about prior information of road surface, a structure preserving model is designed to explore the continuous clues in the spatial level. Experimental results on two benchmark datasets show that the proposed method achieves the state-of-the-art performance, compared with previous approaches for road extraction. Code will be made available on https://github.com/MrStrangerYang/SDUNet.

Extracting Urban Road Footprints from Airborne LiDAR Point Clouds with PointNet++ and Two-Step Post-Processing

In this paper, a novel framework for the automatic extraction of road footprints from airborne LiDAR point clouds in urban areas is proposed. The extraction process consisted of three phases: The first phase is to extract road points by using the deep learning model PointNet++, where the features of the input data include not only those selected from raw LiDAR points, such as 3D coordinate values, intensity, etc., but also the digital number (DN) of co-registered images and generated geometric features to describe a strip-like road. Then, the road points from PointNet++ were post-processed based on graph-cut and constrained triangulation irregular networks, where both the commission and omission errors were greatly reduced. Finally, collinearity and width similarity were proposed to estimate the connection probability of road segments, thereby improving the connectivity and completeness of the road network represented by centerlines. Experiments conducted on the Vaihingen data show that the proposed framework outperformed others in terms of completeness and correctness; in addition, some narrower residential streets with 2 m width, which have normally been neglected by previous studies, were extracted. The completeness and the correctness of the extracted road points were 84.7% and 79.7%, respectively, while the completeness and the correctness of the extracted centerlines were 97.0% and 86.3%, respectively.

BSIRNet: A Road Extraction Network with Bidirectional Spatial Information Reasoning

Automatic extraction of road information from remote sensing images is widely used in many fields, such as urban planning and automatic navigation. However, due to interference from noise and occlusion, the existing road extraction methods can easily lead to road discontinuity. To solve this problem, a road extraction network with bidirectional spatial information reasoning (BSIRNet) is proposed, in which neighbourhood feature fusion is used to capture spatial context dependencies and expand the receptive field, and an information processing unit with a recurrent neural network structure is used to capture channel dependencies. BSIRNet enhances the connectivity of road information through spatial information reasoning. Using the public Massachusetts road dataset and Wuhan University road dataset, the superiority of the proposed method is verified by comparing its results with those of other models.

EUNetMTL: multitask joint learning for road extraction from high-resolution remote sensing images

ABSTRACT Road extraction based on high-resolution remote sensing images has become a major research hotspot. However, the traditional pixel-level semantic segmentation method lacks road continuity details, resulting in road extractions containing many fragmented road segments and, thus, incorrect segmentation results. To address these problems, we propose the EfficientUNet multitask joint learning (EUNetMTL) model – an end-to-end model based on semantic segmentation and orientation learning. Firstly, EUNetMTL uses an encoder–decoder network structure based on EfficientNet-B4 for feature extraction, which provides it with superior feature extraction accuracy. Secondly, the decoder is modified by adding dilated convolution to enlarge the receptive field. Finally, the model incorporates an orientation learning decoding branch, which solves the discontinuity problem in road extraction by sharing the encoding with the segmentation task branch. The results of experiments conducted on the DeepGlobe and SpaceNet road datasets show that the proposed model achieves high-quality segmentation results that are superior to those of current state-of-the-art methods.

FMWDCT: Foreground Mixup Into Weighted Dual-Network Cross Training for Semisupervised Remote Sensing Road Extraction

With the development of deep learning, the application of automatic road extraction has achieved great success. However, the main challenge is how to make full use of a large number of unlabeled images to improve segmentation models and how alleviate sample imbalance in road extraction tasks. In this paper, we propose a novel semi-supervised remote sensing road extraction approach is refined as Foreground Mixup into Weighted Dual-network Cross Training (FMWDCT), which combines labeled images with unlabeled images to extract road from remote sensing images. FMWDCT is composed of Dual-network Cross Training (DCT) and Foreground Pasting (FP). DCT is a new semi-supervised training method and FP is an effective data perturbation method for road extraction. We firstly paste the foreground pixels obtained from labeled images into unlabeled images to produce mixed input images. The mixed pseudo labels are then generated by a combination of high-confidence predictions from the augmented network and labeled masks. Finally, the mixed pseudo labels are used to guide another adversarial basic network for cross training, and this basic network is used to smoothly update the augmented network that corresponds to it. The proposed FMWDCT effectively solve the overfitting problem and imbalance problem of positive and negative sample in the case of a few labeled training samples. We demonstrate the effectiveness of our method on three road extraction datasets, and achieve better performance with few labeled data. Extensive experiments show that the proposed semi-supervised method can learn latent information from the unlabeled data to improve performance.

RoadSeg-CD: A Network With Connectivity Array and Direction Map for Road Extraction From SAR Images

Road extraction from synthetic aperture radar (SAR) images has attracted much attention in the field of remote sensing image processing. General road extraction algorithms, affected by shadows of buildings and trees, are prone to producing fragmented road segments. To improve the accuracy and completeness of road extraction, we propose a neural network-based algorithm which takes the connectivity and direction features of roads into consideration, named RoadSeg-CD. It consists of two branches: one is the main branch for road segmentation; the other is the auxiliary branch for learning road directions. In the main branch, a connectivity array is designed to utilize local contextual information and construct a connectivity loss based on the predicted probabilities of neighboring pixels. In the auxiliary branch, we proposed a novel road direction map, which is used for learning the directions of roads. The two branches are connected by specific feature fusion process, and the output from the main branch is taken as the road extraction result. Experiments on real radar images are implemented to validate the effectiveness of our method. The experimental results demonstrate that our method can obtain more continuous and more complete roads than several state-of-the-art road extraction algorithms.

Road Extraction From Remote Sensing Images in Wildland–Urban Interface Areas

In this letter, we address the problem of road extraction in Wildland–urban interface (WUI) areas. In recent years, with the great success of convolutional neural networks (CNNs) in various vision-related tasks, researchers have developed many CNN-based methods for road extraction on remote sensing images. Nevertheless, these methods mostly treat road extraction as a binary classification problem on semantic labeling. In WUI areas, the road is narrower and tends to be occluded by trees, which may result in the serious discontinuous problem of inferred road maps. To address this issue, we propose transforming the input representation of the binary classification map into a continuous signed distance map. In this way, our model is forced to predict the continuous distance representations and, thus, improve the spatial continuities of inferred roads. In addition, a real-value regression task is designed to train along with the original binary classification task to generate spatially continuous and semantically accurate road maps. Then, we conduct experiments on the public Massachusetts road data set and a homemade data set collected from Yajishan Mountain, Beijing, China. Finally, our proposed method achieves intersection-over-unions (IoUs) of 64.11% and 65.92% for the Massachusetts and WUI-Yajishan data sets, respectively, without any postprocessing. In addition, the ablation analysis shows that introducing the regression task on the proposed signed distance representation can effectively alleviate the problem of discontinuous road prediction. Furthermore, comparing with the state-of-the-art methods demonstrates the superiority of our method for road extraction in WUI areas.

Cascaded Multi-Task Road Extraction Network for Road Surface, Centerline, and Edge Extraction

Road extraction from very high-resolution (VHR) remote sensing imagery remains a huge challenge, due to the shadows and occlusions of trees and buildings. Such complex backgrounds result in deep networks often producing fragmented roads with poor connectivity. Road extraction has three typical tasks: road surface segmentation (SS), centerline extraction (CE), and edge detection (ED), which are conducted in a wide range of real applications. Also, the three tasks have a symbiotic relationship, i.e., the road SS determines the location of the centerline and edges, and the CE and ED can allow the generation of more continuous road surfaces. However, most of the previous works have completed these three tasks separately, without exploiting the symbiotic relationship between them to boost the road connectivity. In this article, in order to improve road connectivity, a cascaded multitask (CasMT) road extraction framework for simultaneously extracting the road surface, centerline, and edges is proposed. In the proposed framework, topology-aware learning is applied to capture the long-distance topological relationships, and hard example mining (HEM) loss is employed to focus more on hard samples, to further enhance the road completeness. Extensive experiments were conducted on the DeepGlobe road dataset and a large-scale road dataset (called the LSCC dataset) from the three Chinese cities of Beijing, Shanghai, and Wuhan. The experimental results obtained on the public DeepGlobe dataset demonstrate that the proposed CasMT framework can significantly outperform the current state-of-the-art method. Moreover, the generalization capability of the model was verified on the LSCC dataset, where the proposed CasMT framework achieved the best performance in the average path length similarity (APLS) road topology metric, which further confirms the superiority of the proposed framework.

Exploring Label Probability Sequence to Robustly Learn Deep Convolutional Neural Networks for Road Extraction With Noisy Datasets

Recently, crowdsourced geographic data have provided a cost-effective approach to learn deep convolutional neural networks (DCNNs) for road extraction from remote sensing images. However, crowdsourced datasets often suffer from label noise and include error labels that can affect the performance of DCNN-based road extraction methods. Thus, we propose a novel sequence deep learning (SDL) framework that introduces label probability sequence to correct error labels and robustly learn DCNNs for road extraction. The label probability sequence is obtained by front-end developed DCNNs, which can provide valuable true label information to alleviate the adverse effect of label noise. Specifically, to pursue high effectiveness of the label correction process, an adaptive label correction method is proposed to exploit the distance between the true label distribution hidden in label probability sequence and the possible true label distribution. In addition, based on the uncertainty of label probability sequence, a noise correction loss function is introduced to robustly learn DCNNs on the improved dataset produced by the adaptive label correction. By repeating these two steps in each epoch, we can learn efficiently DCNN-based road extraction methods under noisy datasets. Experimental studies on two real-world noisy road datasets show that the proposed method can help to overcome label noise issue for DCNNs and achieve excellent performance in road extraction.