In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
530 
DETECTION OF HIGHWAYS IN HIGH RESOLUTION IMAGES USING MATHEMATICAL 
MORPHOLOGY TECHNIQUES 
A. S. Ishikawa 3 , E.A. Silva b 
a,b UNESP - Sao Paulo State University, 305 Roberto Simonsen St., Brazil, 19060-900 
a aline.ishikawa@gmail.com 
b erivaldo@fct.unesp.br 
KEYWORDS: mathematical morphology, cartographic feature, updating, digital image processing, remote sensing, high resolution 
images 
ABSTRACT: 
This paper seeks to apply routine for highways detection through the mathematical morphology tools in high resolution image. The 
Mathematical Morphology theory consists of describing structures geometric presents quantitatively in the image (objectives or features). 
This explains the use of the Mathematical Morphology in this work. As high resolution images will be used, the largest difficulty in the 
highways detection process is the presence of trees and automobiles in the borders tracks. Like this, for the obtaining of good results 
through the use of morphologic tools was necessary to choose the structuring element appropriately to be used in the functions. Through 
the appropriate choice of the morphologic operators and structuring elements it was possible to detect the highways tracks. The linear 
feature detection using mathematical morphology techniques, can contribute in cartographic applications, as cartographic products 
updating. 
1. INTRODUCTION 
Cartographic documents in different scales are fundamental tools 
to know the environment in way local, regional and global, and to 
help in the socket of necessary decisions for the planning and 
administration of resources. The current situation of the world 
cartography indicates that, for many areas of the planet, don't exist 
cartographic documents that they supply information with 
appropriate quality, being this a so much common problem for 
developed nations as for developing countries (Estes; Mooneyhan, 
1994). 
The Brazilian Cartography presents great deficiency in 
cartographic products updating. This happens due to the existent 
emptiness, incorrect scales, products that aren’t adapted to the 
established quality pattern or for lack of an effective politics gone 
back to the territory mapping. Aiming at to reduce that mapping 
lack in great scales, the cartographic community is addressing 
efforts in the search of alternative methodologies so that the 
cartographic products can be updated with larger speed, efficiency 
and smaller cost. This process is of great importance for the 
obtaining of reliable and economically viable cartographic bases, 
without the need to accomplish conventional procedures, that you 
are costly and slow. 
A fundamental factor in the process of cartographic updating is 
the changes detection, which according to Colwell (1974) it 
consists of the modifications recognition happened in the 
terrestrial surface through the time. Among the techniques of 
changes detection, it was chosen for this paper the integrated use 
of Remote Sensing products and Digital Image Processing (DPI) 
techniques. Among the several tools of DPI, it was chosen the 
Mathematical Morphology (MM) for the interest features 
detection in orbital images. The MM is a technique very used and 
researched in Universities and Research Centers of everyone, 
tends as base to study the sign geometry, being a powerful tool for 
images analysis, mainly where geometric aspects are relevant. 
Considering the country wide national cartography deficiency, 
this work seeks to contribute for the automation of the 
Cartography field through the integrated utilization of Remote 
Sensing data and Digital Image Processing techniques. The main 
objective of this work was to apply a routine for highways 
detection in high resolution images (QuickBird) through the 
Mathematical Morphology tools, with views to the cartographic 
updating. 
2. MAIN BODY 
2.1 Mathematical Morphology 
The Mathematical Morphology (MM) was originally developed 
by Matheron and Serra at the École des Mines in Paris. It is a set- 
theoretic method of image analysis providing a quantitative 
description of geometrical structures. 
According to Soille (1999), MM can be defined as theory for the 
analysis of spatial structures. It is called of morphology, because it 
consists at analysing the shape and form of objects. It is 
mathematical because the analysis bases on set theory, integral 
geometry, and lattice algebra. Goutsias and Heijmans (2000) 
mention that, the main idea of the MM is to analyze the shape of 
objects in an image by probing the image with a small geometric 
template (e.g., line segment, disc and square) known as the 
structuring element. 
2.1.1 Structuring Element 
Façon (1996) defines the structuring element as a completely 
defined and known set (form and size). The basic principle of the 
mathematical morphology lives in the operation of sets among the 
image tests and the structuring element known, obtaining the 
relative information to the geometry and the topology of that set.