x
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B7, 2012
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia
ROAD CLASSIFICATION AND CONDITION DETERMINATION USING
HYPERSPECTRAL IMAGERY
M. Mohammadi
Department of Geomatics, Computer Science and Mathematics, University of Applied Sciences Stuttgart
SchellingstraBe 24, D-70174 Stuttgart, Germany — mohammadi, 2025( yahoo.com
KEY WORDS: Hyperspectral, Urban, ALK vector data, Classification, Condition
ABSTRACT:
Hyperspectral data has remarkable capabilities for automatic identification and mapping of urban surface materials because of its high
spectral resolution. It includes a wealth of information which facilitates an understanding of the ground material properties. For
identification of road surface materials, information about their relation to hyperspectral sensor measurements is needed. In this study an
approach for classification of road surface materials using hyperspectral data is developed. The condition of the road surface materials, in
particular asphalt is also investigated. Hyperspectral data with 4m spatial resolution of the city of Ludwigsburg, Germany consisting of
125 bands (wavelength range of 0.4542um to 2.4846 um) is used. Different supervised classification methods such as spectral angle
mapper are applied based on a spectral library established from field measurements and in-situ inspection. It is observed that using the
spectral angle mapper approach with regions of interest is helpful for road surface material identification. Additionally, spectral features
are tested using their spectral functions in order to achieve better classification results. Spectral functions such as mean and standard
deviation are suitable for discriminating asphalt, concrete and gravel. Different asphalt conditions (good, intermediate and bad) are
distinguished using the spectral functions such as mean and image ratio. The mean function gives reliable results. Automatisierte
Liegenschaftskarte (ALK) vector data for roads is integrated in order to confine the analysis to roads. Reliable reference spectra are useful
in evaluation of classification results for spectrally similar road surface materials. The classification results are assessed using
orthophotos and field visits information.
1. INTRODUCTION AND RELATED RESEARCH
Comprehensive information about road networks as one of the
transportation features is helpful for assessment and planning of
transportation (routing). Retrieval of road information such as
road surface material and pavement type condition is one of the
essential issues in urban areas. This is done with either traditional
surveying or remote sensing (RS) (Zhang and Couloigner,
2004).The former needs more labour and is more time consuming
in comparison to the latter. Hyperspectral imagery, also known as
imaging spectrometry, is the acquisition of data in many narrow,
contiguous spectral bands (Goetz et al., 1985). It provides more
detailed information in comparison to other remote sensing
techniques. Different chemical materials such as asphalt and
gravel by their corresponding physical (absorption, albedo,
reflectivity etc) properties can be derived on a very detailed level
from the hyperspectral imagery. This characteristic is helpful in
discrimination and extraction of urban area objects especially
those with similar spectral properties. Road surface materials can
be identified with hyperspectral imagery with less cost compared
to field surveying. Most of the available methods for mapping
roads are either manual or semi-automatic. However, these
approaches are time consuming and expensive. In particular, that
they may involve a lot of field work and interpretation of aerial
imagery from which only
limited information can be acquired. Hyperspectral data has
significant potential in terms of automatic identification of road
surface materials. However, no standard approach for mapping
road surfaces and identifying the condition of road surface
materials exists up to date. Most of the methods that exist were
originally developed for mineral detection. Thus it is a challenge
to use these methods in identifying road surface materials due to
the variation of these materials in relatively small regions in the
case of roads.
In recent times there has been an increase in the demand to find
economical automated methods to extract information from
hyperspectral data due to the fast developments in urban areas.
Noronha et al, (2002) focus on extraction of road centerlines,
detecting pavement condition and developing a spectral library.
The analysis of hyperspectral data is done using MultiSpec
software and maximum likelihood classification is applied.
Bhattacharyya distance is used for separability analysis between
road materials and roof types. For better discrimination of roofs
and roads, an object-oriented image classification technique is
used. This technique tries to analyse the homogenous image object
rather than independent pixels. A multispectral sensor (IKONOS)
is used as an ancillary sensor. Comparison of configuration for
urban target separation in spectrometry and multispectral remote
sensing shows that some of the features are not determined in the
latter. This is due to the broadness and location of the bands.
Moreover, its broad band channels do not resolve small-scale
spectral absorption features which are unique for several built up
areas. According to the separability analysis results, concrete and
