Road Extraction Research Articles

Consistency Self-Training Semi-Supervised Method for Road Extraction from Remote Sensing Images

Road extraction techniques based on remote sensing image have significantly advanced. Currently, fully supervised road segmentation neural networks based on remote sensing images require a significant number of densely labeled road samples, limiting their applicability in large-scale scenarios. Consequently, semi-supervised methods that utilize fewer labeled data have gained increasing attention. However, the imbalance between a small quantity of labeled data and a large volume of unlabeled data leads to local detail errors and overall cognitive mistakes in semi-supervised road extraction. To address this challenge, this paper proposes a novel consistency self-training semi-supervised method (CSSnet), which effectively learns from a limited number of labeled data samples and a large amount of unlabeled data. This method integrates self-training semi-supervised segmentation with semi-supervised classification. The semi-supervised segmentation component relies on an enhanced generative adversarial network for semantic segmentation, which significantly reduces local detail errors. The semi-supervised classification component relies on an upgraded mean-teacher network to handle overall cognitive errors. Our method exhibits excellent performance with a modest amount of labeled data. This study was validated on three separate road datasets comprising high-resolution remote sensing satellite images and UAV photographs. Experimental findings showed that our method consistently outperformed state-of-the-art semi-supervised methods and several classic fully supervised methods.

A High-Resolution Remote Sensing Road Extraction Method Based on the Coupling of Global Spatial Features and Fourier Domain Features

Remote sensing road extraction based on deep learning is an important method for road extraction. However, in complex remote sensing images, different road information often exhibits varying frequency distributions and texture characteristics, and it is usually difficult to express the comprehensive characteristics of roads effectively from a single spatial domain perspective. To address the aforementioned issues, this article proposes a road extraction method that couples global spatial learning with Fourier frequency domain learning. This method first utilizes a transformer to capture global road features and then applies Fourier transform to separate and enhance high-frequency and low-frequency information. Finally, it integrates spatial and frequency domain features to express road characteristics comprehensively and overcome the effects of intra-class differences and occlusions. Experimental results on HF, MS, and DeepGlobe road datasets show that our method can more comprehensively express road features compared with other deep learning models (e.g., Unet, D-Linknet, DeepLab-v3, DCSwin, SGCN) and extract road boundaries more accurately and coherently. The IOU accuracy of the extracted results also achieved 72.54%, 55.35%, and 71.87%.

Rural Road Extraction in Xiong'an New Area of China Based on the RC-MSFNet Network Model.

High-resolution remote sensing imagery, reaching meter or sub-meter levels, provides essential data for extracting and identifying road information. However, rural roads are often narrow, elongated, and have blurred boundaries, with textures that resemble surrounding environments such as construction sites, vegetation, and farmland. These features often lead to incomplete extraction and low extraction accuracy of rural roads. To address these challenges, this study introduces the RC-MSFNet model, based on the U-Net architecture, to enhance rural road extraction performance. The RC-MSFNet model mitigates the vanishing gradient problem in deep networks by incorporating residual neural networks in the downsampling stage. In the upsampling stage, a connectivity attention mechanism is added after dual convolution layers to improve the model's ability to capture road completeness and connectivity. Additionally, the bottleneck section replaces the traditional dual convolution layers with a multi-scale fusion atrous convolution module to capture features at various scales. The study focuses on rural roads in the Xiong'an New Area, China, using high-resolution imagery from China's Gaofen-2 satellite to construct the XARoads rural road dataset. Roads were extracted from the XARoads dataset and DeepGlobe public dataset using the RC-MSFNet model and compared with some models such as U-Net, FCN, SegNet, DeeplabV3+, R-Net, and RC-Net. Experimental results showed that: (1) The proposed method achieved precision (P), intersection over union (IOU), and completeness (COM) scores of 0.8350, 0.6523, and 0.7489, respectively, for rural road extraction in Xiong'an New Area, representing precision improvements of 3.8%, 6.78%, 7.85%, 2.14%, 0.58%, and 2.53% over U-Net, FCN, SegNet, DeeplabV3+, R-Net, and RC-Net. (2) The method excelled at extracting narrow roads and muddy roads with unclear boundaries, with fewer instances of omission or false extraction, demonstrating advantages in complex rural terrain and areas with indistinct road boundaries. Accurate rural road extraction can provide valuable reference data for urban development and planning in the Xiong'an New Area.

Multi-feature fusion and multi-attention deep network for enhancing road extraction in remote sensing images

ABSTRACT Road detection in remote sensing (RS) images plays a critical role in applications ranging from urban planning to autonomous navigation systems. However, accurate road extraction remains a challenging task due to the presence of textual-similar objects that can be visually confused with roads, and shadows that can obscure road features. For this regard, we present a novel end-to-end network that leverages multi-feature fusion and multi-attention. Multi-level dual residual blocks are introduced to capture multi-scale and multi-level road features. A channel attention feature fusion module fuses road features from the decoder while suppressing coarse-grain noises. Moreover, a difference channel feature fusion module filters out the interfering texture-similar features and irrelevant background features. Subsequently, a spatial channel feature enhancement module is designed to identify the mis-segmented regions and reclassify these pixels. We conducted extensive experiments on diverse datasets, demonstrating the effectiveness of our approach in improving road detection accuracy, especially in areas with shadows and textual-similar objects. The results indicate that our approach outperforms state-of-the-art methods in handling textual-similar objects and shadow-occluded roads, making it a valuable contribution to the field of road extraction in remote sensing images.

Road extraction from remote sensing images by combining attention and context fusion

Aiming at the problem that the objects in remote sensing images are complex, and the roads are long, thin, continuously distributed and easily blocked, a road extraction model for remote sensing images combining attention and context fusion (ACFD-LinkNet) is proposed. The model is based on the D-LinkNet network. First, a strip attention module is used after the last convolutional layer of the D-LinkNet network encoder to enhance the feature extraction ability of roads of different scales, better capture the global features of the road, and capture the long-distance information of the road; secondly, a context fusion module (CFM) is proposed and added to the feature transfer part of the network codec to predict the road connection between adjacent pixels, fuse the road information between different levels of the context, and solve the problem of obstacles blocking the road connection; finally, the cross entropy loss function and Dice loss function of the improved model are set with multi-loss function hyperparameter weight distribution to solve the imbalance of positive and negative samples in the data set, and the best segmentation accuracy is obtained by adjusting the weight ratio. Experiments were conducted on DeepGlobe and CHN6-CUG datasets, and the comprehensive index F1 value reached 86.76%、92.12%, which was improved compared with the D-LinkNet model . 3.96%、1.13%In addition, compared with Unet, Deeplabv3+, A 2-FPN and other networks, it has the best performance.

Multi-Modal Contextualization of Trajectory Data for Advanced Analysis

SummaryRising amounts of generated geospatial data, either trajectory-like tracking data, raster-like imagery, or vector-like mappings as in OpenStreetMap (OSM), grow the need for multi-modal algorithmic analysis. Existing machine-learning-based algorithms contradictly mainly focus on image and textual input representations and cannot deal with other modes of geospatial data. Therefore, we propose a novel method to contextualize vector-like trajectory data with surrounding data to create easy-to-be-analyzed image-like representations. Our approach includes the proposition of a chase-cam-like scanline over space according to the trajectory’s speed and possibly smoothed orientation. Thereby, surrounding pixels in the vicinity of the trajectory points are accumulated along the scanline and are combined into a visual representation of the trajectory. To show the potential effects of our work, we predict traffic regulations for trajectory sections in the vehicle speed dataset based on our proposed trajectory-based sampling of orthophotos in the same region. This proposes a new way of using multi-modal data sources (trajectories and airborne imagery) to extract road metadata.

Improved Road Extraction Models through Semi-Supervised Learning with ACCT

Improving the performance and reducing the training cost of road extraction models in the absence of samples is important for updating road maps. Despite the success of recent road extraction models on standard datasets, they often fail to perform when applied to new datasets or real-world scenarios where labeled samples are not available. In this paper, our focus diverges from the typical quest to pinpoint the optimal road extraction model or evaluate generalization prowess across models. Instead, we propose a method called Asymmetric Consistent Co-Training (ACCT) to train existing road extraction models faster and make them perform better in new scenarios lacking samples. ACCT uses two models with different structures and a supervision module to enhance accuracy through mutual learning. Labeled and unlabeled images are processed by both models to generate road maps from different perspectives. The supervision module ensures consistency between predictions by computing losses based on labeling status. ACCT iteratively adjusts parameters using unlabeled data, improving generalization. Empirical evaluations show that ACCT improves IoU by 2.79% to 10.26% using only 1/8 of the labeled data compared to fully supervised methods. It also reduces parameters by over 49% compared to state-of-the-art semi-supervised methods while maintaining similar accuracy. These results highlight the potential of leveraging large amounts of unlabeled data to enhance road extraction models as data acquisition technology advances.

Narrow road extraction from high-resolution remote sensing images: SWGE-Net and MSIF-Net

ABSTRACT Accurate and complete road network extraction plays a critical role in urban planning, street navigation, and emergency response. At present, narrow roads are a main feature in most public road datasets. However, the continuity and boundary completeness of the extraction results for these narrow roads are relatively poor, due to their varied shapes, uneven spatial distribution, and the presence of various interfering elements. To address these issues, this study introduces a novel network, the Self-weighted Global Context Road Extraction Network (SWGE-Net), which integrates a dilate block and an improved coordinate attention mechanism to effectively capture the complex details and spatial information of narrow roads. Furthermore, most public road training datasets often lack labels for very narrow roads, this omission leads to poor extraction results for these roads in test datasets. In order to further improve the extraction capability for unlabeled, extremely narrow roads, this study introduces another network called the Multi-scale Information Fusion Road Extraction Network (MSIF-Net), which uses the same encoders as SWGE-Net and has a special module for merging information at different scales. This module, with a dilate block and pyramid pooling-based decoder, makes the network better at recognizing and combining features of different sizes. Experimental results indicate that SWGE-Net outperforms the baseline network with road IoU scores of 71.57% and 60.67% on the DeepGlobe and CHN6-CUG road datasets, respectively an improvement of 18.51% and 5.40%. Meanwhile, MSIF-Net not only exceeds the baseline in road IoU scores for both datasets, but also achieves the best performance in extracting unlabeled, extremely narrow roads in qualitative experiments.

Unsupervised Multispectral Gaussian Mixture Model-Based Framework for Road Extraction

Unsupervised Multispectral Gaussian Mixture Model-Based Framework for Road Extraction

Retraction Note: Extraction of road using soft computing techniques

Retraction Note: Extraction of road using soft computing techniques

Visual Navigation of Caged Chicken Coop Inspection Robot Based on Road Features.

The speed and accuracy of navigation road extraction and driving stability affect the inspection accuracy of cage chicken coop inspection robots. In this paper, a new grayscale factor (4B-3R-2G) was proposed to achieve fast and accurate road extraction, and a navigation line fitting algorithm based on the road boundary features was proposed to improve the stability of the algorithm. The proposed grayscale factor achieved 92.918% segmentation accuracy, and the speed was six times faster than the deep learning model. The experimental results showed that at the speed of 0.348 m/s, the maximum deviation of the visual navigation was 4 cm, the average deviation was 1.561 cm, the maximum acceleration was 1.122 m/s2, and the average acceleration was 0.292 m/s2, with the detection number and accuracy increased by 21.125% and 1.228%, respectively. Compared with inertial navigation, visual navigation can significantly improve the navigation accuracy and stability of the inspection robot and lead to better inspection effects. The visual navigation system proposed in this paper has better driving stability, higher inspection efficiency, better inspection effect, and lower operating costs, which is of great significance to promote the automation process of large-scale cage chicken breeding and realize rapid and accurate monitoring.

Large-scale integration of remotely sensed and GIS road networks: A full image-vector conflation approach based on optimization and deep learning

Road networks play an important role in the sustainable development of human society. Conventionally, there are two sources of road data acquisition: road extraction from Remote Sensing (RS) imagery and GIS based map production. Each method has its limitations. The RS road extraction methods are primarily raster-based and the extracted roads are not directly usable in GIS due to their fragmented and noisy nature, while vector-based methods cannot utilize rich raster information. Further more, the vector and raster data can have discrepancies for various reasons. Efficient road data production requires an image-vector conflation process that can match and combine raster and vector-based road data automatically.In this study, we propose a full image-vector conflation framework that directly integrates image and vector road data by appropriately transforming extracted roads from imagery and establishing a match relation between these roads and a credible target GIS road dataset. Based on analyzing these match relations, we propose new metrics for measuring the degree of agreement between the raster and vector road data. The proposed framework combines state-of-the-art deep learning methods for image segmentation and optimization-based models for object matching. We prepared a large-scale high-resolution road dataset covering two counties in Kansas, US. Using trained models from one of the two counties, we were able to extract road segments in the other county and match them to the TIGER/Line roads.Our experiments show that conventional performance metrics for road extraction (e.g. IoU) are insufficient for measuring the degree of agreement between image and vector roads as they are pixel-based and are too sensitive to spatial displacement. Instead, the newly defined vector-based agreement metrics are needed for image-vector conflation purposes. Experiments show that, by the vector-based metrics, nearly 90% of GIS road lengths in the study area were extracted and over 90% of extracted roads matched the target GIS roads. The new framework streamlines raster-vector conflation of roads and can potentially expedite relevant geospatial analyses regarding change detection, disaster monitoring and GIS data production, among others.

Intelligent road extraction from high resolution remote sensing images based on optimized SVM

Road recognition and extraction based on high-resolution remote sensing images is currently a hot issue at the forefront of image processing and other disciplines, and its results not only help to enrich geographic information, but also have important application value in national defence construction and other aspects. For the intelligent extraction of roads in high-resolution remote sensing images, the study firstly preprocesses these images, and then extracts the roads by using variational function and support vector machine. As the extraction results are not satisfactory enough, the 3D wavelet transform technique is used to extract the spectral features of roads in the frequency domain, combined with the texture in the time domain, and optimized the support vector machine in order to extract the road images more accurately. The outcomes denote that in contrast with other methods, the optimized method can have optimal correctness and quality while guaranteeing higher integrity. The satellite road image of Chengdu A industrial park is rich in extracted information, having the highest completeness of 94.23%, correctness of 60.73%, and quality of 59.42%, and the satellite road image of Chengdu B industrial park is similarly rich in extracted information, having the highest completeness of 83.47%, correctness of 85.31%, and quality of 72.85%. The highest value of CPU usage for road extraction was 2000 kb at 37 s, and the average value of CPU usage for the whole 60 s was 500 kb. It shows that the research method has good results in high resolution remote sensing image road extraction.

Insights into the Effects of Tile Size and Tile Overlap Levels on Semantic Segmentation Models Trained for Road Surface Area Extraction from Aerial Orthophotography

Studies addressing the supervised extraction of geospatial elements from aerial imagery with semantic segmentation operations (including road surface areas) commonly feature tile sizes varying from 256 × 256 pixels to 1024 × 1024 pixels with no overlap. Relevant geo-computing works in the field often comment on prediction errors that could be attributed to the effect of tile size (number of pixels or the amount of information in the processed image) or to the overlap levels between adjacent image tiles (caused by the absence of continuity information near the borders). This study provides further insights into the impact of tile overlaps and tile sizes on the performance of deep learning (DL) models trained for road extraction. In this work, three semantic segmentation architectures were trained on data from the SROADEX dataset (orthoimages and their binary road masks) that contains approximately 700 million pixels of the positive “Road” class for the road surface area extraction task. First, a statistical analysis is conducted on the performance metrics achieved on unseen testing data featuring around 18 million pixels of the positive class. The goal of this analysis was to study the difference in mean performance and the main and interaction effects of the fixed factors on the dependent variables. The statistical tests proved that the impact on performance was significant for the main effects and for the two-way interaction between tile size and tile overlap and between tile size and DL architecture, at a level of significance of 0.05. We provide further insights and trends in the predictions of the extensive qualitative analysis carried out with the predictions of the best models at each tile size. The results indicate that training the DL models on larger tile sizes with a small percentage of overlap delivers better road representations and that testing different combinations of model and tile sizes can help achieve a better extraction performance.

Learning by counterexamples in remote sensing image classification: a case study for road extraction

ABSTRACT The classification of multi- and hyper-spectral imagery is essential for several remote sensing applications. However, several challenges still hinder the accuracy and efficiency of learning algorithms. In diverse remote sensing applications, the high spectral heterogeneity between and within classes is often an issue that adds uncertainties to the classification. Most classifiers do not consider the influence of non-target classes and variations in the spectral similarity level between classes during classification. This paper presents a novel learning strategy for remote sensing image classification, which exploits non-target classes to enhance the generation of classification masks and the performance of supervised methods for specific classes of interest. It produces multiple classification instances, based on the incorporation of counterexample information into the training data set, in order to refine the separation of classes of interest in the available feature space. The experimental results demonstrate the potential of the proposed counterexample assisted learning for road extraction, using high spatial-resolution imagery data.

A Lightweight CNN Based on Axial Depthwise Convolution and Hybrid Attention for Remote Sensing Image Dehazing

Hazy weather reduces contrast, narrows the dynamic range, and blurs the details of the remote sensing image. Additionally, color fidelity deteriorates, causing color shifts and image distortion, thereby impairing the utility of remote sensing data. In this paper, we propose a lightweight remote sensing-image-dehazing network, named LRSDN. The network comprises two tailored, lightweight modules arranged in cascade. The first module, the axial depthwise convolution and residual learning block (ADRB), is for feature extraction, efficiently expanding the convolutional receptive field with little computational overhead. The second is a feature-calibration module based on the hybrid attention block (HAB), which integrates a simplified, yet effective channel attention module and a pixel attention module embedded with an observational prior. This joint attention mechanism effectively enhances the representation of haze features. Furthermore, we introduce a novel method for remote sensing hazy image synthesis using Perlin noise, facilitating the creation of a large-scale, fine-grained remote sensing haze image dataset (RSHD). Finally, we conduct both quantitative and qualitative comparison experiments on multiple publicly available datasets. The results demonstrate that the LRSDN algorithm achieves superior dehazing performance with fewer than 0.1M parameters. We also validate the positive effects of the LRSDN in road extraction and land cover classification applications.

Dirty road extraction from GF-2 images by semi-supervised deep learning method for arid and semiarid regions of southern Mongolia

ABSTRACT The uncontrolled proliferation of natural roads in arid regions has exacerbated regional land degradation and desertification, presenting substantial challenges to their accurate mapping owing to their dynamic and obscure features. Moreover, the high cost of data annotation restricts the availability of comprehensively labelled datasets, which are essential for advanced remote sensing processing and natural road detection. This study dedicated to implement a semi-supervised deep learning method for dirty road extraction in southern Mongolia. A new thematic semantic segmentation dataset of natural roads was established firstly to address scarcity of annotation datasets this region. A semi-supervised UniMatch structure was designed consequently. Operating with high-resolution GaoFen-2 images, this approach minimises the need for extensive manual annotation, achieving an IOU of 73.51% and MIOU of 86.37%. This method significantly reduces labour and time costs associated with manual and fully supervised methods. These observations provide a valuable data source and methodology for addressing natural road expansion in arid regions. They can aid governments in evaluating transportation infrastructure in remote areas, and analysing dirty road traffic impact on environment.

Road extraction from remote sensing images based on a multi-scale asymmetric dual attention mechanism

ABSTRACT Aiming at the problems of road fracture and detail loss caused by not considering the geometric features in the road extraction method. We proposed an encoder-decoder architecture based on a multi-scale asymmetric dual attention mechanism. Firstly, A multi-scale convolution block in the shape of ‘Union Jack’ is designed. It includes symmetric convolution and asymmetric convolution along horizontal, vertical, left diagonal, and right diagonal spatial directions, and a multi-scale dilated convolution for extracting features of different scales. Remote dependence relationships are highly converged by using it, and road fracture problems caused by occlusion can be solved effectively. Secondly, a directional dual attention mechanism is proposed, which consists of directional channel attention using strip pooling and a directional spatial attention mechanism using asymmetric convolution along left diagonal and right diagonal spatial directions. It can use the directivity of asymmetric convolution to allocate attention mechanism adaptively in attention mechanism, and effectively avoid the road detail loss problem. Finally, we conducted corresponding experiments on the DeepGlobe and Ottawa road datasets, and the experimental results are superior to the current state-of-the-art methods.

Local Evaluation of Large-scale Remote Sensing Machine Learning-generated Building and Road Dataset: The Case of Rwanda

AbstractAccurate and up-to-date building and road data are crucial for informed spatial planning. In developing regions in particular, major challenges arise due to the limited availability of these data, primarily as a result of the inherent inefficiency of traditional field-based surveys and manual data generation methods. Importantly, this limitation has prompted the exploration of alternative solutions, including the use of remote sensing machine learning-generated (RSML) datasets. Within the field of RSML datasets, a plethora of models have been proposed. However, these methods, evaluated in a research setting, may not translate perfectly to massive real-world applications, attributable to potential inaccuracies in unknown geographic spaces. The scepticism surrounding the usefulness of datasets generated by global models, owing to unguaranteed local accuracy, appears to be particularly concerning. As a consequence, rigorous evaluations of these datasets in local scenarios are essential for gaining insights into their usability. To address this concern, this study investigates the local accuracy of large RSML datasets. For this evaluation, we employed a dataset generated using models pre-trained on a variety of samples drawn from across the world and accessible from public repositories of open benchmark datasets. Subsequently, these models were fine-tuned with a limited set of local samples specific to Rwanda. In addition, the evaluation included Microsoft’s and Google’s global datasets. Using ResNet and Mask R‑CNN, we explored the performance variations of different building detection approaches: bottom-up, end-to-end, and their combination. For road extraction, we explored the approach of training multiple models on subsets representing different road types. Our testing dataset was carefully designed to be diverse, incorporating both easy and challenging scenes. It includes areas purposefully chosen for their high level of clutter, making it difficult to detect structures like buildings. This inclusion of complex scenarios alongside simpler ones allows us to thoroughly assess the robustness of DL-based detection models for handling diverse real-world conditions. In addition, buildings were evaluated using a polygon-wise comparison, while roads were assessed using network length-derived metrics.Our results showed a precision (P) of around 75% and a recall (R) of around 60% for the locally fine-tuned building model. This performance was achieved in three out of six testing sites and is considered the lowest limit needed for practical utility of RSML datasets, according to the literature. In contrast, comparable results were obtained in only one out of six sites for the Google and Microsoft datasets. Our locally fine-tuned road model achieved moderate success, meeting the minimum usability threshold in four out of six sites. In contrast, the Microsoft dataset performed well on all sites. In summary, our findings suggest improved performance in road extraction, relative to building extraction tasks. Moreover, we observed that a pipeline relying on a combination of bottom-up and top-down segmentation, while leveraging open global benchmark annotation dataset as well as a small number of samples for fine-tuning, can offer more accurate RSML datasets compared to an open global dataset. Our findings suggest that relying solely on aggregated accuracy metrics can be misleading. According to our evaluation, even city-level derived measures may not capture significant variations in performance within a city, such as lower accuracy in specific neighbourhoods. Overcoming the challenges of complex areas might benefit from exploring alternative approaches, including the integration of LiDAR data, UAV images, aerial images or using other network architectures.

RIRNet: A Direction-Guided Post-Processing Network for Road Information Reasoning

Road extraction from high-resolution remote sensing images (HRSIs) is one of the tasks in image analysis. Deep convolutional neural networks have become the primary method for road extraction due to their powerful feature representation capability. However, roads are often obscured by vegetation, buildings, and shadows in HRSIs, resulting in incomplete and discontinuous road extraction results. To address this issue, we propose a lightweight post-processing network called RIRNet in this study, which include an information inference module and a road direction inference task branch. The information inference module can infer spatial information relationships between different rows or columns of feature images from different directions, effectively inferring and repairing road fractures. The road direction inference task branch performs the road direction prediction task, which can constrain and promote the road extraction task, thereby indirectly enhancing the inference ability of the post-processing model and realizing the optimization of the initial road extraction results. Experimental results demonstrate that the RIRNet model can achieve an excellent post-processing effect, which is manifested in the effective repair of broken road segments, as well as the handling of errors such as omission, misclassification, and noise, proving the effectiveness and generalization of the model in post-processing optimization.