Spatio-temporal Road Detection from Aerial Imagery using CNNs

Luque, Belen; Rubió, Josep Ramon Morros; Ruiz-Hidalgo, Javier

doi:10.5220/0006128904930500

Cited by 6 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Remote sensing optical data obtained from UAVs or airborne/spaceborne platforms are mainly used for the extraction of roads, as described in several previous studies [ 18 , 19 , 20 , 21 , 22 ]. Meanwhile, synthetic aperture radar (SAR) data, especially high-resolution X-band SAR datasets, focus on the quality of the road.…”

Section: Methodsmentioning

confidence: 99%

Development of Nationwide Road Quality Map: Remote Sensing Meets Field Sensing

Karimzadeh

Matsuoka

2021

Sensors

View full text Add to dashboard Cite

In this study, we measured the in situ international roughness index (IRI) for first-degree roads spanning more than 1300 km in East Azerbaijan Province, Iran, using a quarter car (QC). Since road quality mapping with in situ measurements is a costly and time-consuming task, we also developed new equations for constructing a road quality proxy map (RQPM) using discriminant analysis and multispectral information from high-resolution Sentinel-2 images, which we calibrated using the in situ data on the basis of geographic information system (GIS) data. The developed equations using optimum index factor (OIF) and norm R provide a valuable tool for creating proxy maps and mitigating hazards at the network scale, not only for primary roads but also for secondary roads, and for reducing the costs of road quality monitoring. The overall accuracy and kappa coefficient of the norm R equation for road classification in East Azerbaijan province are 65.0% and 0.59, respectively.

show abstract

Section: Methodsmentioning

confidence: 99%

Development of Nationwide Road Quality Map: Remote Sensing Meets Field Sensing

Karimzadeh

Matsuoka

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…Similarly, Hu et al [36] take a FCN-like approach to extract features from the last convolutional layer at multiple scales and encode them into global image features through feature coding techniques. In [37], this task is approached from a temporal perspective by modifying SegNet to extract roads from low resolution videos captured by Unmanned Aerial Vehicles (UAVs). The model uses feature maps from the corresponding encoder stage to upsample and adds an iterative algorithm to process the temporal dimension.…”

Section: Related Workmentioning

confidence: 99%

A Deep Learning-Based Solution for Large-Scale Extraction of the Secondary Road Network from High-Resolution Aerial Orthoimagery

et al. 2020

View full text Add to dashboard Cite

Secondary roads represent the largest part of the road network. However, due to the absence of clearly defined edges, presence of occlusions, and differences in widths, monitoring and mapping them represents a great effort for public administration. We believe that recent advancements in machine vision allow the extraction of these types of roads from high-resolution remotely sensed imagery and can enable the automation of the mapping operation. In this work, we leverage these advances and propose a deep learning-based solution capable of efficiently extracting the surface area of secondary roads at a large scale. The solution is based on hybrid segmentation models trained with high-resolution remote sensing imagery divided in tiles of 256 × 256 pixels and their correspondent segmentation masks, resulting in increases in performance metrics of 2.7–3.5% when compared to the original architectures. The best performing model achieved Intersection over Union and F1 scores of maximum 0.5790 and 0.7120, respectively, with a minimum loss of 0.4985 and was integrated on a web platform which handles the evaluation of large areas, the association of the semantic predictions with geographical coordinates, the conversion of the tiles’ format and the generation of geotiff results compatible with geospatial databases.

show abstract

“…In [24], this challenge is approached from a spatio-temporal perspective by adding a temporal processing block on top of existing networks, achieving accuracy levels over 90%. In [25], the authors proposed a model based on VGGNet to learn the features of road boundaries by integrating RGB images street scenes, the semantic contour and the location in a neural network for autonomous driving applications.…”

Section: Related Workmentioning

confidence: 99%

A Framework Based on Nesting of Convolutional Neural Networks to Classify Secondary Roads in High Resolution Aerial Orthoimages

et al. 2020

View full text Add to dashboard Cite

Remote sensing imagery combined with deep learning strategies is often regarded as an ideal solution for interpreting scenes and monitoring infrastructures with remarkable performance levels. In addition, the road network plays an important part in transportation, and currently one of the main related challenges is detecting and monitoring the occurring changes in order to update the existent cartography. This task is challenging due to the nature of the object (continuous and often with no clearly defined borders) and the nature of remotely sensed images (noise, obstructions). In this paper, we propose a novel framework based on convolutional neural networks (CNNs) to classify secondary roads in high-resolution aerial orthoimages divided in tiles of 256 × 256 pixels. We will evaluate the framework’s performance on unseen test data and compare the results with those obtained by other popular CNNs trained from scratch.

show abstract

Spatio-temporal Road Detection from Aerial Imagery using CNNs

Cited by 6 publications

References 11 publications

Development of Nationwide Road Quality Map: Remote Sensing Meets Field Sensing

Development of Nationwide Road Quality Map: Remote Sensing Meets Field Sensing

A Deep Learning-Based Solution for Large-Scale Extraction of the Secondary Road Network from High-Resolution Aerial Orthoimagery

A Framework Based on Nesting of Convolutional Neural Networks to Classify Secondary Roads in High Resolution Aerial Orthoimages

Contact Info

Product

Resources

About