Massachusetts Roads Dataset Research Articles

RemainNet: Explore Road Extraction from Remote Sensing Image Using Mask Image Modeling

Road extraction from a remote sensing image is a research hotspot due to its broad range of applications. Despite recent advancements, achieving precise road extraction remains challenging. Since a road is thin and long, roadside objects and shadows cause occlusions, thus influencing the distinguishment of the road. Masked image modeling reconstructs masked areas from unmasked areas, which is similar to the process of inferring occluded roads from nonoccluded areas. Therefore, we believe that mask image modeling is beneficial for indicating occluded areas from other areas, thus alleviating the occlusion issue in remote sensing image road extraction. In this paper, we propose a remote sensing image road extraction network named RemainNet, which is based on mask image modeling. RemainNet consists of a backbone, image prediction module, and semantic prediction module. An image prediction module reconstructs a masked area RGB value from unmasked areas. Apart from reconstructing original remote sensing images, a semantic prediction module of RemainNet also extracts roads from masked images. Extensive experiments are carried out on the Massachusetts Roads dataset and DeepGlobe Road Extraction dataset; the proposed RemainNet improves 0.82–1.70% IoU compared with other state-of-the-art road extraction methods.

Advanced road extraction using CNN-based U-Net model and satellite imagery

Road extraction from high-resolution remote sensing imagery (HRRSI) is a challenging task due to low spectral variation and the presence of complex background elements, such as the shadows of buildings and trees. Many techniques have struggled to maintain proper edges and boundaries while retaining the geometric features necessary for accurate non-linear road extraction. In this paper, we address these issues by proposing a deep learning approach that utilizes a fine-tuned U-Net model with a modified feature space in the basic U-Net architecture for semantic segmentation. We also incorporate the BRISQUE preprocessing technique to improve the performance. We experimented with a subset of 200 high-quality images from the Massachusetts roads dataset, prioritized based on their rank. Due to computer memory constraints, the images were resized to 256 by 256 pixels. The proposed method produced the accuracy of 95.45%.

Remote Sensing Image Road Extraction Network Based on MSPFE-Net

Road extraction is a hot task in the field of remote sensing, and it has been widely concerned and applied by researchers, especially using deep learning methods. However, many models using convolutional neural networks ignore the attributes of roads, and the shape of the road is banded and discrete. In addition, the continuity and accuracy of road extraction are also affected by narrow roads and roads blocked by trees. This paper designs a network (MSPFE-Net) based on multi-level strip pooling and feature enhancement. The overall architecture of MSPFE-Net is encoder-decoder, and this network has two main modules. One is a multi-level strip pooling module, which aggregates long-range dependencies of different levels to ensure the connectivity of the road. The other module is the feature enhancement module, which is used to enhance the clarity and local details of the road. We perform a series of experiments on the dataset, Massachusetts Roads Dataset, a public dataset. The experimental data showed that the model in this paper was better than the comparison models.

Traffic Analysis Through Deep-Learning-Based Image Segmentation From UAV Streaming

Alongside many traditional as well as novel applications, in the latest years, drones have been widely adopted as remote sensing platforms for road traffic monitoring in urban areas and on highways. The problem of traffic monitoring on Region of Interest (RoI) based on drone imagery is a challenging task, especially when the surveillance drone is constantly moving. In this work, two specific subtasks have been addressed. The goal of the first stage is to predict the RoI in drone imagery of traffic scenes using deep-learning (DL)-based approaches instead of traditional image processing; in this connection, the goal of the second task is to perform vehicle detection on the selected RoI. To ensure diversity and robustness, drone images with different altitudes, positions, and view points have been considered. To achieve these goals, two custom aerial data sets for RoI extraction and detection were built by collecting aerial sequences from flying unmanned aerial vehicles (UAVs) and by transmitting them to the base station leveraging 5G technology. Two different ad hoc DL-based architectures have been designed for the RoI extraction task to maximize the accuracy and inference speed, respectively, and have been evaluated on two different data sets: 1) a custom-built data set and 2) a Massachusetts roads data set. Finally, the models providing the best performance have been combined to further improve the overall results. Experimental tests show that the proposed framework represents a promising solution for drone-based road traffic monitoring in critical areas, exploiting imagery from a variety of viewing angles and altitudes.

CSANet: Cross-Scale Axial Attention Network for Road Segmentation

Road segmentation from remote sensing images is an important task in many applications. However, due to the high density of roads and the complex background, the roads are often occluded by trees. This makes accurate road segmentation a challenge task. Most existing road segmentation networks rely on convolutions with small kernels; however, these methods often cannot obtain satisfying results because the long-range dependencies are not captured and the intrinsic relationships between feature maps at different scales are not fully exploited. In this paper, a deep neural network based on a cross-scale axial attention mechanism is proposed to address this problem. This model enables low-resolution features to aggregate global contextual information from high-resolution features. Among them, the axial attention mechanism realizes global attention by using vertical and horizontal attention sequentially. With this strategy, the dense long-range dependencies can be captured with extremely low computational cost. The cross-scale mechanism enables the model to effectively combine the high-resolution fine-grained features and the low-resolution coarse-grained features. The proposed method enables the network to propagate the information without losing details. Our method achieves IoUs of 58.98 and 65.28 on the Massachusetts Roads dataset and DeepGlobe dataset and outperforms other methods.

Using AAEHS-Net as an Attention-Based Auxiliary Extraction and Hybrid Subsampled Network for Semantic Segmentation.

Semantic segmentation based on deep learning has undergone remarkable advancements in recent years. However, due to the neglect of the shallow features, the problems of inaccurate segmentation have persisted. To address this issue, a semantic segmentation network-attention-based auxiliary extraction and hybrid subsampled network (AAEHS-Net) is suggested in this study. To extract more deep information and the shallow features, the complementary and enhanced extraction module (CEEM) is utilized by the network. As a result, the edge segmentation of the model is improved. Moreover, to reduce the loss of features, a hybrid subsampled module (HSM) is introduced. Meanwhile, global max pool and global avg pool module (GAGM) is designed as an attention module to enhance the features with global and important information and maintain feature continuity. The proposed AAEHS-Net is evaluated on three datasets: the aerial drone image dataset, the Massachusetts roads dataset, and the Massachusetts buildings dataset. On the three datasets, AAEHS-Net achieves 1.15%, 0.88%, and 2.1% higher accuracy than U-Net, reaching 90.12%, 96.23%, and 95.15%, respectively. At the same time, our proposed network has obtained the best values for all evaluation metrics in three datasets compared to the currently popular algorithms.

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.

Split Depth-Wise Separable Graph-Convolution Network for Road Extraction in Complex Environments From High-Resolution Remote-Sensing Images

Road information from high-resolution remote-sensing images is widely used in various fields, and deep-learning-based methods have effectively shown high road-extraction performance. However, for the detection of roads sealed with tarmac, or covered by trees in high-resolution remote-sensing images, some challenges still limit the accuracy of extraction: 1) large intraclass differences between roads and unclear interclass differences between urban objects, especially roads and buildings; 2) roads occluded by trees, shadows, and buildings are difficult to extract; and 3) lack of high-precision remote-sensing datasets for roads. To increase the accuracy of road extraction from high-resolution remote-sensing images, we propose a split depth-wise (DW) separable graph convolutional network (SGCN). First, we split DW-separable convolution to obtain channel and spatial features, to enhance the expression ability of road features. Thereafter, we present a graph convolutional network to capture global contextual road information in channel and spatial features. The Sobel gradient operator is used to construct an adjacency matrix of the feature graph. A total of 13 deep-learning networks were used on the Massachusetts roads dataset and nine on our self-constructed mountain road dataset, for comparison with our proposed SGCN. Our model achieved a mean intersection over union (mIOU) of 81.65% with an F1-score of 78.99% for the Massachusetts roads dataset, and an mIOU of 62.45% with an F1-score of 45.06% for our proposed dataset. The visualization results showed that SGCN performs better in extracting covered and tiny roads and is able to effectively extract roads from high-resolution remote-sensing images.

DDU-Net: Dual-Decoder-U-Net for Road Extraction Using High-Resolution Remote Sensing Images

Extracting roads from high-resolution remote sensing images (HRSIs) is vital in a wide variety of applications, such as autonomous driving, path planning, and road navigation. Due to the long and thin shape as well as the shades induced by vegetation and buildings, small-sized roads are more difficult to discern. In order to improve the reliability and accuracy of small-sized road extraction when roads of multiple sizes coexist in an HRSI, an enhanced deep neural network model termed Dual-Decoder-U-Net (DDU-Net) is proposed in this paper. Motivated by the U-Net model, a small decoder is added to form a dual-decoder structure for more detailed features. In addition, we introduce the dilated convolution attention module (DCAM) between the encoder and decoders to increase the receptive field as well as to distill multi-scale features through cascading dilated convolution and global average pooling. The convolutional block attention module (CBAM) is also embedded in the parallel dilated convolution and pooling branches to capture more attention-aware features. Extensive experiments are conducted on the Massachusetts Roads dataset with experimental results showing that the proposed model outperforms the state-of-the-art DenseUNet, DeepLabv3+ and D-LinkNet by 6.5%, 3.3%, and 2.1% in the mean Intersection over Union (mIoU), and by 4%, 4.8%, and 3.1% in the F1 score, respectively. Both ablation and heatmap analyses are presented to validate the effectiveness of the proposed model.

ConDinet++: Full-Scale Fusion Network Based on Conditional Dilated Convolution to Extract Roads From Remote Sensing Images

Extracting roads from aerial images is an issue that has attracted much attention. Using semantic segmentation methods to extract roads often faces the problem of narrow and occluded roads. In this letter, we propose a network called ConDinet++, which improves the general codec architecture. In the encoder part, the VGG16 with pretraining parameters is utilized for the feature extraction. In the decoder part, we perform a feature fusion mechanism on the full-scale feature map. In order to improve the ability of the network to extract and integrate semantic information and further increase the receptive field, we recommend adopting the conditional dilated convolution blocks (CDBs) in the encoder, and each CDB consists of a group of cascaded conditional dilated convolutions. More importantly, the designed codec architecture can adjust the number of convolutions and the parameters of the convolution kernel according to the input data. For a slender area like a road, which occupies a small area in the picture, we use the joint loss function and introduce the joint loss of Lovasz loss and cross-entropy loss to avoid the segmentation model having a serious bias caused by highly unbalanced object sizes between roads and background. The proposed method was tested on two public datasets Massachusetts Roads Dataset and Mini DeepGlobe Road Extraction Challenge. Compared with some previous semantic segmentation networks, the proposed ConDinet++ achieved the best values of recall, F-score, and mIoU.

An enhanced global context-aware network for road extraction from high-resolution imagery

ABSTRACT Currently, the automation of road segmentation in high-resolution satellite imagery is inadequate, and a part of the road network information is affected by the loss of spatial features and lack of global context information. To overcome this, we propose a deep convolutional neural network called the enhanced global context-aware network (EGC-Net). The EGC-Net employs a newly enhanced global context-aware module, which adds convolution kernels of different scales based on atrous spatial pyramid pooling (ASPP). In addition, the feature aggregation and self-attention modules further improve the accuracy of road segmentation. Our experimental results show that the performance of EGC-Net is superior to that of state-of-the-art methods. When tested using the Massachusetts Roads Dataset, the EGC-Net achieved an F1-score (evaluation metric for the harmonic mean of the precision and recall metrics) of 0.8234, accuracy of 0.9466, and the intersection of the prediction and ground truth regions over their union (IoU) of 0.8195. Overall, EGC-Net performs better than other deep learning algorithms under different evaluation indexes. The ablation study indicates that the IoU score of our enhanced global context-aware module is higher than that of ASPP by 0.0207, and each component is required to obtain the best road detection results.

Deep Feature-Review Transmit Network of Contour-Enhanced Road Extraction From Remote Sensing Images

The acquisition of road information from remote sensing images is of significant value with regard to intelligent transportation research. This study focuses on enhancing the contour-learning ability to mitigate the phenomenon of fragmented road segments and missing connections in road extraction. A novel Deep Feature-Review (FR) Transmit Network (TransNet) is proposed to review and facilitate the flow of contour features into an encoder network. Meanwhile, multiscale features are linked via a bridge between the encoder and the decoder. Compared with the state-of-the-art models such as fully convolutional network (FCN), SegNet, DeepLabv3, D-LinkNet, spatial consistency-FCN, and generative adversarial network (GAN), the proposed network achieves better overall performance for the Massachusetts Roads data set, with accuracy, precision, recall, and mean intersection-over-union (IoU) scores of 97.48%, 83.72%, 78.13%, and 0.6286%, respectively. For the DeepGlobe Road Extraction data set, the proposed network outperforms FCN, SegNet, DeepLabv3, D-LinkNet, and Deep TransNet, achieving accuracy, precision, recall, and mean IoU scores of 98.70%, 87.30%, 81.15%, and 0.7244%, respectively. Overall, these experiments indicate that the proposed network can effectively address the phenomenon of fragmented road segments and poor connectivity in remote sensing images, indicating its potential for utilization in practical intelligent transportation scenarios.

UnetDVH-Linear: Linear Feature Segmentation by Dilated Convolution with Vertical and Horizontal Kernels.

Linear feature extraction is crucial for special objects in semantic segmentation networks, such as slot marking and lanes. The objects with linear characteristics have global contextual information dependency. It is very difficult to capture the complete information of these objects in semantic segmentation tasks. To improve the linear feature extraction ability of the semantic segmentation network, we propose introducing the dilated convolution with vertical and horizontal kernels (DVH) into the task of feature extraction in semantic segmentation networks. Meanwhile, we figure out the outcome if we put the different vertical and horizontal kernels on different places in the semantic segmentation networks. Our networks are trained on the basis of the SS dataset, the TuSimple lane dataset and the Massachusetts Roads dataset. These datasets consist of slot marking, lanes, and road images. The research results show that our method improves the accuracy of the slot marking segmentation of the SS dataset by 2%. Compared with other state-of-the-art methods, our UnetDVH-Linear (v1) obtains better accuracy on the TuSimple Benchmark Lane Detection Challenge with a value of 97.53%. To prove the generalization of our models, road segmentation experiments were performed on aerial images. Without data argumentation, the segmentation accuracy of our model on the Massachusetts roads dataset is 95.3%. Moreover, our models perform better than other models when training with the same loss function and experimental settings. The experiment result shows that the dilated convolution with vertical and horizontal kernels will enhance the neural network on linear feature extraction.

Road Extraction from Very-High-Resolution Remote Sensing Images via a Nested SE-Deeplab Model

Automatic road extraction from very-high-resolution remote sensing images has become a popular topic in a wide range of fields. Convolutional neural networks are often used for this purpose. However, many network models do not achieve satisfactory extraction results because of the elongated nature and varying sizes of roads in images. To improve the accuracy of road extraction, this paper proposes a deep learning model based on the structure of Deeplab v3. It incorporates squeeze-and-excitation (SE) module to apply weights to different feature channels, and performs multi-scale upsampling to preserve and fuse shallow and deep information. To solve the problems associated with unbalanced road samples in images, different loss functions and backbone network modules are tested in the model’s training process. Compared with cross entropy, dice loss can improve the performance of the model during training and prediction. The SE module is superior to ResNext and ResNet in improving the integrity of the extracted roads. Experimental results obtained using the Massachusetts Roads Dataset show that the proposed model (Nested SE-Deeplab) improves F1-Score by 2.4% and Intersection over Union by 2.0% compared with FC-DenseNet. The proposed model also achieves better segmentation accuracy in road extraction compared with other mainstream deep-learning models including Deeplab v3, SegNet, and UNet.

Road Extraction from Unmanned Aerial Vehicle Remote Sensing Images Based on Improved Neural Networks.

Roads are vital components of infrastructure, the extraction of which has become a topic of significant interest in the field of remote sensing. Because deep learning has been a popular method in image processing and information extraction, researchers have paid more attention to extracting road using neural networks. This article proposes the improvement of neural networks to extract roads from Unmanned Aerial Vehicle (UAV) remote sensing images. D-Linknet was first considered for its high performance; however, the huge scale of the net reduced computational efficiency. With a focus on the low computational efficiency problem of the popular D-LinkNet, this article made some improvements: (1) Replace the initial block with a stem block. (2) Rebuild the entire network based on ResNet units with a new structure, allowing for the construction of an improved neural network D-Linknetplus. (3) Add a 1 × 1 convolution layer before DBlock to reduce the input feature maps, reducing parameters and improving computational efficiency. Add another 1 × 1 convolution layer after DBlock to recover the required number of output channels. Accordingly, another improved neural network B-D-LinknetPlus was built. Comparisons were performed between the neural nets, and the verification were made with the Massachusetts Roads Dataset. The results show improved neural networks are helpful in reducing the network size and developing the precision needed for road extraction.

Automatic Road Segmentation from High Resolution Satellite Images Using Encoder-Decoder Network

Road network segmentation from high resolution satellite imagery have profound applications in remote sensing. They facilitate for transportation, GPS navigation and digital cartography. Most recent advances in automatic road segmentation leverage the power of networks such as fully convolutional networks and encoder-decoder networks. The main disadvantage with these networks is that they contain deep architectures with large number of hidden layers to account for the lost spatial and localization features. This will add a significant computational overhead. It is also difficult to segment roads from other road-like features. In this paper, we propose a road segmentation architecture with an encoder and two path decoder modules. One path of the decode module approximates the coarse spatial features using upsampling network. The other path uses Atrous spatial pyramid pooling module to extract multi scale context information. Both the decoder paths are combined to fine tune the segmented road network. The experiments on the Massachusetts roads dataset show that our proposed model can produce precise segmentation results than other state-of-the-art models without being computationally expensive.

Light encoder–decoder network for road extraction of remote sensing images

Extracting roads from remote sensing images is an important task in the remote sensing field. We propose an approach of designing a light encoder–decoder network for road extraction. We analyze the relationship between road features and the receptive field of encoder–decoder networks and point out that the light encoder–decoder network can be achieved by controlling its receptive field. Based on this, we design road extraction networks on the architecture of a general encoder–decoder network, according to data specifications. In addition, we propose an adaptive weighted binary cross-entropy loss function to solve the problem of data imbalance in the training process. We validate our approach on the Massachusetts roads dataset and the DeepGlobe road extraction dataset. The experimental results show that our method reduces 98% and 94% of the parameters, respectively, compared with general encoder–decoder networks, whereas the performance of road extraction keeps well. Our approach has fewer parameters and good performance, so it is easier to deploy on mobile platforms.

Road Network Extraction using Atrous Spatial Pyramid Pooling

Road extraction from satellite images has several Applications such as geographic information system (GIS). Having an accurate and up-to-date road network database will facilitate transportation, disaster management and GPS navigation. Most active field of research for automatic extraction of road network involves semantic segmentation using convolutional neural network (CNN). Although they can produce accurate results, typically the models give up performance for accuracy and vice-versa. In this paper, we are proposing architecture for semantic segmentation of road networks using Atrous Spatial Pyramid Pooling (ASPP). The network contains residual blocks for extracting low level features. Atrous convolutions with different dilation rates are taken and spatial pyramid pooling is performed on these features for extracting the spatial information. The low level features from residual blocks are added to the multi scale context information to produce the final segmentation image. Our proposed model significantly reduces the number of parameters that are required to train the model. The proposed model was trained on the Massachusetts roads dataset and the results have shown that our model produces superior results than that of popular state-of-the art models.

Road Network Extraction Using Atrous Spatial Pyramid Pooling

Road extraction from satellite images has several Applications such as geographic information system (GIS). Having an accurate and up-to-date road network database will facilitate transportation, disaster management and GPS navigation. Most active field of research for automatic extraction of road network involves semantic segmentation using convolutional neural network (CNN). Although they can produce accurate results, typically the models give up performance for accuracy and vice-versa. In this paper, we are proposing architecture for semantic segmentation of road networks using Atrous Spatial Pyramid Pooling (ASPP). The network contains residual blocks for extracting low level features. Atrous convolutions with different dilation rates are taken and spatial pyramid pooling is performed on these features for extracting the spatial information. The low level features from residual blocks are added to the multi scale context information to produce the final segmentation image. Our proposed model significantly reduces the number of parameters that are required to train the model. The proposed model was trained on the Massachusetts roads dataset and the results have shown that our model produces superior results than that of popular state-of-the art models.

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.