Detection of linear features in SAR images: application to road network extraction

Abstract

We propose a two-step algorithm for almost unsupervised detection of linear structures, in particular, main axes in road networks, as seen in synthetic aperture radar (SAR) images. The first step is local and is used to extract linear features from the speckle radar image, which are treated as roadsegment candidates. We present two local line detectors as well as a method for fusing information from these detectors. In the second global step, we identify the real roads among the segment candidates by defining a Markov random field (MRF) on a set of segments, which introduces contextual knowledge about the shape of road objects. The influence of the parameters on the road detection is studied and results are presented for various real radar images. Index Terms-Markov random fields (MRF's), road detection, SAR images, statistical properties. NOMENCLATURE Number of looks of the radar image. Amplitude of pixel. Number of pixels in region. Empirical mean of region. Empirical variation coefficient of region. Exact mean-reflected intensity of region. , Exact and empirical contrasts between regions and. Ratio edge detector response between regions and. Ratio line detector (D1) response. Cross-correlation edge detector response between regions and. Cross-correlation line detector (D2) response. Decision threshold for variable. Probability-density function (pdf) of a random variable for value and parameter values. Cumulative distribution function of a random variable for value and parameter values .

Fuzzy fusion techniques for linear features detection in multitemporal SAR images

“…We decided to seek roads that are up to pixels wide, which is a typical maximum response width in ERS-1 images for common roads [12].…”

“…Different methods based on edge detectors [12], [17], [18], statistical [15], [16], Markovian [12], [14], or neural [13] approaches have been proposed in the literature for the extraction of linear structures. But, since the features we are looking for are intrinsically characterized by their shape, it seemed natural to us to use morphological operators [8]- [11].…”

“…It aims at suppressing the false alarms and at reconnecting the detected segments in order to fully reconstruct the road network. Tupin [12], for instance, recently proposed such a reconnection step based on a markovian approach. Nevertheless, the better the local detection is, the better the final reconnection will be.…”

“…These operators are defined as a linear combination of the ranked input data: (12) Depending on the weighting vector ( ), the min and max operators can be considered as particular cases of OWA operators, as well as the arithmetical mean or even the quantified operator introduced by (10) (if , else ). This particular operator will be tested on real data in the following.…”

A framework based on spin glass models for the inference of anatomical connectivity from diffusion‐weighted MR data – a technical review

“…A similar idea was previously proposed to restore a road network from very noisy radar satellite images. 29 …”