Automatic Road Network Extraction Research Articles

Automatic Extraction of Road Networks by Using Teacher–Student Adaptive Structural Deep Belief Network and Its Application to Landslide Disaster

Automatic Extraction of Road Networks by Using Teacher–Student Adaptive Structural Deep Belief Network and Its Application to Landslide Disaster

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.

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.

Extraction of road using soft computing techniques

One of the most challenging research topics is extraction of road network automatically from remotely sensed imagery. The roads are considered as backbone and important modes of transportation. For efficient automatic road network extraction, ground points and building points extracted were integrated and it was used with multistage road network extraction model. For road network extraction, a lane detection system is considered as a more important element. From the lane marking in the roads, the roads are easily detected from the remotely sensed images. In order to detect the lanes more effectively, B-snake with fuzzy C-means (FCM) clustering is proposed in this paper. The input image is divided into equal halves by using FCM clustering algorithm. Then, the lower part is binarized by analyzing RGB channel and it is analyzed by using B-snake model because it contains more information about road lane than upper part of the image. Then, the lane detection problem is formulated by finding the set of lane model control points. A good initial position to the B-snake lane model is provided by introducing Canny/Hough estimation of vanishing points. Moreover, the external force field for lane detection is constructed by using gradient vector flow. The correspondence between B-snake and the real edge image is determined by using minimum mean square error which is presented to find out the parameters of road model iteratively. The detected lane points are fused with the road and ground points by using constraint satisfaction neural network-complementary information integration. Moreover, the road and non-road regions are separated by using region part segmentation. Finally, the medial-axis-transform-based hypothesis verification is used to automatically complete the road network extraction framework. The performance of the existing road network extraction, ground point-based road extraction model, building and ground point-based road extraction model and proposed building, ground and lane point-based road extraction model is analyzed in terms of correctness and completeness.

Road Centerline Extraction from Very-High-Resolution Aerial Image and LiDAR Data Based on Road Connectivity

The road networks provide key information for a broad range of applications such as urban planning, urban management, and navigation. The fast-developing technology of remote sensing that acquires high-resolution observational data of the land surface offers opportunities for automatic extraction of road networks. However, the road networks extracted from remote sensing images are likely affected by shadows and trees, making the road map irregular and inaccurate. This research aims to improve the extraction of road centerlines using both very-high-resolution (VHR) aerial images and light detection and ranging (LiDAR) by accounting for road connectivity. The proposed method first applies the fractal net evolution approach (FNEA) to segment remote sensing images into image objects and then classifies image objects using the machine learning classifier, random forest. A post-processing approach based on the minimum area bounding rectangle (MABR) is proposed and a structure feature index is adopted to obtain the complete road networks. Finally, a multistep approach, that is, morphology thinning, Harris corner detection, and least square fitting (MHL) approach, is designed to accurately extract the road centerlines from the complex road networks. The proposed method is applied to three datasets, including the New York dataset obtained from the object identification dataset, the Vaihingen dataset obtained from the International Society for Photogrammetry and Remote Sensing (ISPRS) 2D semantic labelling benchmark and Guangzhou dataset. Compared with two state-of-the-art methods, the proposed method can obtain the highest completeness, correctness, and quality for the three datasets. The experiment results show that the proposed method is an efficient solution for extracting road centerlines in complex scenes from VHR aerial images and light detection and ranging (LiDAR) data.

Automatic Extraction of Road Networks from GPS Traces

We propose a point segmentation and grouping method to generate road maps from GPS traces. First, we present a progressive point cloud segmentation algorithm based on Total Least Squares (tls) line fitting. Second, we group topologically connected point clusters by the point’s orientation and cluster’s spatial proximity, where the topological relationship is generated using Hidden Markov Model (hmm) map matching. Finally, we refine the intersections of roads so that their geometrical and topological relationships are consistent with each other. Experimental results show that our algorithm is robust to noises and the generated road network has a high accuracy in terms of geometry and topology. Compare to the representative algorithms; the results of our new algorithm have a higher F-measure score for different matching thresholds.

Automatic Extraction of Road Networks from GPS Traces

We propose a point segmentation and grouping method to generate road maps from GPS traces. First, we present a progressive point cloud segmentation algorithm based on Total Least Squares (<small>TLS</small>) line fitting. Second, we group topologically connected point clusters by the point's orientation and cluster's spatial proximity, where the topological relationship is generated using Hidden Markov Model (<small>HMM</small>) map matching. Finally, we refine the intersections of roads so that their geometrical and topological relationships are consistent with each other. Experimental results show that our algorithm is robust to noises and the generated road network has a high accuracy in terms of geometry and topology. Compare to the representative algorithms; the results of our new algorithm have a higher F-measure score for different matching thresholds.

FPGA IMPLEMENTATION OF ROAD NETWORK EXTRACTION USING MORPHOLOGICAL OPERATOR

In the remote sensing analysis, automatic extraction of road network from satellite or aerial images is the most needed approach for efficient road database creation, refinement, and updating. Mathematical morphology is a tool for extracting the features of an image that are useful in the representation and description of region shape. Morphological operator plays a significant role in the extraction of road network from satellite images. Most of the image processing algorithms need to handle large amounts of data, high repeatability, and general software is relatively slow to implement, so the system cannot achieve real-time requirements. In this paper, field programmable gate array (FPGA) architecture designed for automatic extraction of road centerline using morphological operator is proposed. Based on simulation and implementation, results are discussed in terms of register transfer level (RTL) design, FPGA editor and resource estimation. For synthesis and implementation of the above architecture, Spartan 3 XC3S400TQ144-4 device is used. The hardware implementation results are compared with software implementation results. The performance of proposed method is evaluated by comparing the results with ground truth road map as reference data and performance measures such as completeness, correctness and quality are calculated. In the software imple-mentation, the average value of completeness, correctness, and quality of various images are 90%, 96%, and 87% respectively. In the hardware implementation, the average value of completeness, correctness, and quality of various images are 87%, 94%, and 85% respectively. These measures prove that the proposed work yields road network very closer to reference road map.

ROAD NETWORK EXTRACTION FROM DSM BY MATHEMATICAL MORPHOLOGY AND REASONING

The objective of this research is the automatic extraction of the road network in a scene of the urban area from a high resolution digital surface model (DSM). Automatic road extraction and modeling from remote sensed data has been studied for more than one decade. The methods vary greatly due to the differences of data types, regions, resolutions et al. An advanced automatic road network extraction scheme is proposed to address the issues of tedium steps on segmentation, recognition and grouping. It is on the basis of a geometric road model which describes a multiple-level structure. The 0-dimension element is intersection. The 1-dimension elements are central line and side. The 2-dimension element is plane, which is generated from the 1-dimension elements. The key feature of the presented approach is the cross validation for the three road elements which goes through the entire procedure of their extraction. The advantage of our model and method is that linear elements of the road can be derived directly, without any complex, non-robust connection hypothesis. An example of Japanese scene is presented to display the procedure and the performance of the approach.

ROAD NETWORK EXTRACTION FROM DSM BY MATHEMATICAL MORPHOLOGY AND REASONING

Abstract. The objective of this research is the automatic extraction of the road network in a scene of the urban area from a high resolution digital surface model (DSM). Automatic road extraction and modeling from remote sensed data has been studied for more than one decade. The methods vary greatly due to the differences of data types, regions, resolutions et al. An advanced automatic road network extraction scheme is proposed to address the issues of tedium steps on segmentation, recognition and grouping. It is on the basis of a geometric road model which describes a multiple-level structure. The 0-dimension element is intersection. The 1-dimension elements are central line and side. The 2-dimension element is plane, which is generated from the 1-dimension elements. The key feature of the presented approach is the cross validation for the three road elements which goes through the entire procedure of their extraction. The advantage of our model and method is that linear elements of the road can be derived directly, without any complex, non-robust connection hypothesis. An example of Japanese scene is presented to display the procedure and the performance of the approach.

Connected component-based technique for automatic extraction of road centerline in high resolution satellite images

In remote sensing analysis, automatic extraction of road network from satellite or aerial images can be a most needed approach for efficient road database creation, refinement, and updating. In this paper, a method for automatic extraction of road centerline from high resolution satellite image is proposed. The proposed work consists of three stages: segmentation of road region, connected component-based operations used to extract the connected road component from segmented road region, and removal of unwanted non-road pixels using morphological operation. The proposed algorithm is implemented on various satellite images, and the results are given in this work. The performance of the proposed method is evaluated by comparing the results with ground truth road map as reference data, and performance measures such as completeness, correctness, and quality are calculated. The average value of completeness, correctness, and quality of various images are 90%, 96%, and 87%, respectively. These measures prove that the proposed work yields road network very closer to reference road map. The proposed method yields very good result for noisy image also, and it proved that the proposed method is insensitive for noise. Performance measures of the proposed work are compared with other methods, and this comparison proves that the proposed method yields very good results than other methods.

Road Extraction from Satellite Images Using Unscented Kalman Filter and Gauss-Hermite Kalman Filter

To many geographic systems (GIS) application scheme such as urban planning and navigation, updating road network database is critical problem. Rapidly changing urban environments accelerate the need for frequent updates or revisions of road network databases. With the advent of high-resolution satellite images, there has been a resurgence of research interest in road extraction techniques. However, due to the extreme complexity of an urban scene, automatic road network extraction continues to be challenging research topic. In this paper, we have proposed a road map extraction system with two efficient filters using satellite images. Here, Unscented Kalman filter (UKF) is used in combination with Gauss-Hermite Kalman Filter (GHKF) to trace and identify various connected road paths and to avoid obstacles under diverse conditions. Unscented Kalman filter (UKF) component is responsible for tracing axis coordinates of a road region until it comes to a severe obstacle or an intersection. Then, the Gauss-Hermite Kalman Filter (GHKF) module takes the control of the road extraction process and regains track of the road or possibly road branches on the other side of a road junction or obstacle. From the results, we ensure that the proposed road extraction technique outperformed the existing approach by achieving the accuracy of 98.452% in cluster 10.

Automatic Road Extraction based on Normalized Cuts and Level set Methods

Automatic road network extraction based on high resolution satellite image for urban planning holds great potential for significant reduction of database development/updating cost and turnaround time. Satellite remote sensing has been recognized worldwide as an effective technology for the monitoring and mapping the urban development. Two approaches for road network extraction for an urban region have been proposed. When an image is considered in original form it is difficult and computationally expensive to extract roads due to presence of other road-like features with straight edges. Hence roads are first extracted as elongated regions by removing bright regions (that mostly represent the buildings, parking lots and other open spaces), non-linear noise segments are removed median filtering (based upon the fact that road networks constitute large number of small linear structures).The roads are then modeled as boundaries and are extracted using Level set and Normalized cuts methods .Finally The extracted roads are overlayed on the original image. The experimental results show that these approaches are efficient in extracting road segments in urban region from high resolution satellite images. Evaluation of the results carried out by comparing the level set and normalized cuts results with manually extracted reference data. The methods were applied on the high resolution IKONOS image of urban area of Hobart, Australia.

Road Network Extraction and Intersection Detection From Aerial Images by Tracking Road Footprints

In this paper, a new two-step approach (detecting and pruning) for automatic extraction of road networks from aerial images is presented. The road detection step is based on shape classification of a local homogeneous region around a pixel. The local homogeneous region is enclosed by a polygon, called the footprint of the pixel. This step involves detecting road footprints, tracking roads, and growing a road tree. We use a spoke wheel operator to obtain the road footprint. We propose an automatic road seeding method based on rectangular approximations to road footprints and a toe-finding algorithm to classify footprints for growing a road tree. The road tree pruning step makes use of a Bayes decision model based on the area-to-perimeter ratio (the A/P ratio) of the footprint to prune the paths that leak into the surroundings. We introduce a lognormal distribution to characterize the conditional probability of A/P ratios of the footprints in the road tree and present an automatic method to estimate the parameters that are related to the Bayes decision model. Results are presented for various aerial images. Evaluation of the extracted road networks using representative aerial images shows that the completeness of our road tracker ranges from 84% to 94%, correctness is above 81%, and quality is from 82% to 92%.

Optimization of road detection from high-resolution satellite images using texture parameters in neural network classifiers

The aim of road detection is to discriminate between road and background pixels. This discrimination is considered to be the most important stage in automatic road network extraction from satellite imagery. In this paper, neural networks are applied to high-resolution IKONOS and QuickBird images for road detection. This paper has endeavored to optimize the functionality of neural networks using a variety of texture parameters. These parameters had different window sizes and gray level numbers, not only from the source but also from the preclassified image. It was discovered that using texture parameters from a preclassified image accompanied by primary spectral information in reclassifying the source image could improve both road and background detection ability of the neural network. Accuracy assessment parameters were evaluated on several pan-sharpened IKONOS and QuickBird images. The obtained results attest to the efficiency of the proposed method.

Automatic extraction of road networks from map images

AbstractWith the advent of sophisticated and diversified car navigation systems in recent years, it is becoming increasingly important to construct a precise and detailed map database. Among map data, road‐network information is a representation of the connection relations of roads by a graph structure, and is one of the most important kinds of information used in route search and guidance, which is the basic function of car navigation systems. At present, much manpower and time are being spent in the preparation of road‐network information, and it is desirable to reduce the labor required by automation using computers. This paper presents a method in which the road network is automatically extracted from map information. First, the road area is extracted from the map image by stochastic relaxation based on the geometrical properties of the road area and the background area. Then, thinning and piecewise‐linear approximation are applied to the road area to extract the road‐network information, which consists of the graphic structure of nodes and links. In this process, the distortion near junctions produced by thinning is corrected, and the nodes are optimally rearranged so that the road network represents the centers of the roads. Then, precise road‐network information is stably extracted. © 2003 Wiley Periodicals, Inc. Electron Comm Jpn Pt 2, 86(4): 62–72, 2003; Published online in Wiley InterScience (www.interscience. wiley.com). DOI 10.1002/ecjb.10144

15-7 地図画像からの道路ネットワークの自動抽出

15-7 地図画像からの道路ネットワークの自動抽出

Finding Road Seeds in Aerial Images

We describe RoadF, a road finding system designed to drive road tracking programs either for automatic extraction of road networks or for testing and performance evaluation. RoadF utilizes clues in the form of antiparallel edge pairs, and, by using a special algorithm to analyze noisy road center sequences, spans gaps in these clues to construct road hypotheses even when edge continuity is imperfect. As a result long road hypotheses are generated, and the longer a road hypothesis is, the more confidence there is in its being correct. In linking hypothesized road centers, RoadF ignores geometric constraints, leading to obvious mistakes that are later corrected by a smoothness checking phase of the road finding program. Further improvement of results is achieved by linking road hypotheses that are consistent in a geometric sense. The main contribution of this research is the treatment of imperfect continuity. Whereas previous research has centered mainly on improved detection of local road properties, successful detection of imperfect continuity makes utilization of length more effective in pruning away false detections. The technique we presented can be used in conjunction with any of the local road property tests employed in previous work, and it can be expected that it would improve the capability to distinguish between real roads and false alarms.