Method Based on Edge Constraint and Fast Marching for Road Centerline Extraction from Very High-Resolution Remote Sensing Images

Gao, Lipeng; Shi, Wenzhong; Zhang, Miao; Lv, Zhiyong

doi:10.3390/rs10060900

Cited by 25 publications

(19 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…( , , ) = ( − , − , − ) (1) Figure 2. The workflow of Node Layer for Node (1,2) . C, H, W represent channel numbers, heights, and widths, respectively, of the feature maps in level 1.…”

Section: Node Layer and Ultra-hierarchical Sampling Blockmentioning

confidence: 99%

“…Hence, high-resolution remote sensing images are getting easier to acquire. An important use for remote sensing images is extracting and mapping artificial objects, such as buildings [1], roads [2], and vehicles [3] at the pixel-level. Among them, building extraction is the most critical task, and it is commonly applied to monitor the subtle changes in urban areas, urban planning, and estimating the population.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Web-Net: A Novel Nest Networks with Ultra-Hierarchical Sampling for Building Extraction from Aerial Imageries

et al. 2019

View full text Add to dashboard Cite

How to efficiently utilize vast amounts of easily accessed aerial imageries is a critical challenge for researchers with the proliferation of high-resolution remote sensing sensors and platforms. Recently, the rapid development of deep neural networks (DNN) has been a focus in remote sensing, and the networks have achieved remarkable progress in image classification and segmentation tasks. However, the current DNN models inevitably lose the local cues during the downsampling operation. Additionally, even with skip connections, the upsampling methods cannot properly recover the structural information, such as the edge intersections, parallelism, and symmetry. In this paper, we propose the Web-Net, which is a nested network architecture with hierarchical dense connections, to handle these issues. We design the Ultra-Hierarchical Sampling (UHS) block to absorb and fuse the inter-level feature maps to propagate the feature maps among different levels. The position-wise downsampling/upsampling methods in the UHS iteratively change the shape of the inputs while preserving the number of their parameters, so that the low-level local cues and high-level semantic cues are properly preserved. We verify the effectiveness of the proposed Web-Net in the Inria Aerial Dataset and WHU Dataset. The results of the proposed Web-Net achieve an overall accuracy of 96.97% and an IoU (Intersection over Union) of 80.10% on the Inria Aerial Dataset, which surpasses the state-of-the-art SegNet 1.8% and 9.96%, respectively; the results on the WHU Dataset also support the effectiveness of the proposed Web-Net. Additionally, benefitting from the nested network architecture and the UHS block, the extracted buildings on the prediction maps are obviously sharper and more accurately identified, and even the building areas that are covered by shadows can also be correctly extracted. The verified results indicate that the proposed Web-Net is both effective and efficient for building extraction from high-resolution remote sensing images.The building extraction task has drawn the attention of researchers over recent years. Before the common application of deep learning, there were massive machine learning models that tried to handle this task. In general, the pixel-wise labelling model consists of two sub-modules: the feature descriptor for extracting the semantic features from the original images and the pixel-wise classifier for determining the classes of the pixels. Some carefully designed feature descriptors were widely used in early approaches. Tuermer et al. [4] firstly used the histogram of gradient (HOG) feature descriptor in remote sensing for detecting vehicles. The Haar feature is applied in [5] for detecting buildings' outlines and determining the location of buildings' corners. Additionally, Yang et al. [6] applied the Scale-invariant feature transform [7] (SIFT) for classifying objects in remote sensing images. Unlike the artificially designed feature descriptors, the trainable models are the mainstream for the choices of classifiers....

show abstract

Section: Node Layer and Ultra-hierarchical Sampling Blockmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Web-Net: A Novel Nest Networks with Ultra-Hierarchical Sampling for Building Extraction from Aerial Imageries

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The fast marching method (FMM) is a simple and computationally efficient numerical method used for solving boundary problems, such as the motion of a propagating front. Previously, this method has been applied to numerous fields, such as road extraction from satellite imagery [58], path planning [59,60], and medical image segmentation [61]. In this work, we treated the active fire visible in the DNB as a region with a defined boundary, and, by setting a seed point inside the area, FMM could segment the image and label each pixel as inside or outside the object based on a set threshold [62].…”

Section: Viirs-dnb Fire Spot Delineationmentioning

confidence: 99%

Mapping and Tracking Forest Burnt Areas in the Indio Maiz Biological Reserve Using Sentinel-3 SLSTR and VIIRS-DNB Imagery

Chiang

Ulloa

2019

Sensors

View full text Add to dashboard Cite

Wildfires are considered one of the most major hazards and environmental issues worldwide. Recently, Earth observation satellite (EOS) sensors have proven to be effective for wildfire detection, although the quality and usefulness of the data are often hindered by cloud presence. One practical workaround is to combine datasets from multiple sensors. This research presents a methodology that utilizes data of the recently-launched Sentinel-3 sea and land surface temperature radiometer (S3-SLSTR) to reflect its applicability for detecting wildfires. In addition, visible infrared imaging radiometer suite day night band (VIIRS-DNB) imagery was introduced to assure day-night tracking capabilities. The wildfire event in the Indio Maiz Biological Reserve, Nicaragua, during 3–13 April 2018, was the study case. Six S3-SLSTR images were processed to compute spectral indices, such as the normalized difference vegetation index (NDVI), the normalized difference water index (NDWI), and the normalized burn ratio (NBR), to perform image segmentation for estimating the burnt area. The results indicate that 5870.7 ha of forest was affected during the wildfire, close to the 5945 ha reported by local authorities. In this study, the fire expansion was delineated and tracked in the Indio Maiz Biological Reserve using a modified fast marching method on nighttime-sensed temporal VIIRS-DNB. This study shows the importance of S3-SLSRT for wildfire monitoring and how it can be complemented with VIIRS-DNB to track burning biomass at daytime and nighttime.

show abstract

“…Traditional image segmentation algorithms, including the classical Otsu threshold segmentation [21] algorithm, FCM clustering algorithm [22], K-means algorithm [23], and watershed algorithm [24], can only segment grayscale images. However, high-resolution remote sensing images provide high spatial resolution and multiple bands with abundant spectral information, so the multiscale segmentation algorithm is used to segment the images [25]. On the basis of studying the extraction method of road centerline, this paper selects the subvoxel precise skeleton extraction algorithm proposed by Robert Van Uiterta et al to extract the road centerline, and the method is improved by the FMM algorithm, which extracts the centerline of organs efficiently and accurately.…”

Section: Introductionmentioning

confidence: 99%

Road Information Extraction from High-Resolution Remote Sensing Images Based on Road Reconstruction

Zhou

Sun

2019

Remote Sensing

View full text Add to dashboard Cite

Traditional road extraction algorithms, which focus on improving the accuracy of road surfaces, cannot overcome the interference of shelter caused by vegetation, buildings, and shadows. In this paper, we extract the roads via road centerline extraction, road width extraction, broken centerline connection, and road reconstruction. We use a multiscale segmentation algorithm to segment the images, and feature extraction to get the initial road. The fast marching method (FMM) algorithm is employed to obtain the boundary distance field and the source distance field, and the branch backing-tracking method is used to acquire the initial centerline. Road width of each initial centerline is calculated by combining the boundary distance fields, before a tensor field is applied for connecting the broken centerline to gain the final centerline. The final centerline is matched with its road width when the final road is reconstructed. Three experimental results show that the proposed method improves the accuracy of the centerline and solves the problem of broken centerline, and that the method reconstructing the roads is excellent for maintain their integrity.

show abstract

Method Based on Edge Constraint and Fast Marching for Road Centerline Extraction from Very High-Resolution Remote Sensing Images

Cited by 25 publications

References 45 publications

Web-Net: A Novel Nest Networks with Ultra-Hierarchical Sampling for Building Extraction from Aerial Imageries

Web-Net: A Novel Nest Networks with Ultra-Hierarchical Sampling for Building Extraction from Aerial Imageries

Mapping and Tracking Forest Burnt Areas in the Indio Maiz Biological Reserve Using Sentinel-3 SLSTR and VIIRS-DNB Imagery

Road Information Extraction from High-Resolution Remote Sensing Images Based on Road Reconstruction

Contact Info

Product

Resources

About