Laura Keyes 
  
APPLYING COMPUTER VISION TECHNIQUES TO TOPOGRAPHIC OBJECTS 
Laura Keyes, Adam Winstanley 
Department of Computer Science 
National University of Ireland Maynooth 
Co. Kildare, Ireland 
lkeyes @cs.may.ie, Adam. Winstanley @may.ie 
KEY WORDS: shape analysis, shape description, object recognition, Fourier descriptors, moment invariants, 
" Boundary chain-coding, scalar descriptors. 
ABSTRACT 
Automatic structuring (feature coding and object recognition) of topographic data, such as that derived from air survey 
or raster scanning large-scale paper maps, requires the classification of objects such as buildings, roads, rivers, fields 
and railways. The recognition of objects is largely based on the matching of descriptions of shapes. Fourier descriptors, 
moment invariants, boundary chain coding and scalar descriptors are widely used in image processing and computer 
vision to describe and classify shapes. They have been developed to describe shape irrespective of position, orientation 
and scale. The applicability of the above four methods to topographic shapes is described and their usefulness 
evaluated. 
1 INTRODUCTION 
Automatic structuring (feature coding and object recognition) of topographic data, such as that derived from air survey 
or raster scanning large-scale paper maps, requires the classification of objects such as buildings, roads, rivers, fields 
and railways. Shape and context are the main attributes used by humans. Our project combines shape recognition 
techniques developed for computer vision and contextual models derived from statistical language theory to recognise 
objects. This paper describes the measurement of shape to characterise features that will then be used as input into a 
graphical language model. 
Much work has been done in computer vision on the identification and classification of objects within images. 
However, less progress has been made on automating feature extraction and semantic capture in vector graphics. This is 
partly because the low-level graphical content of maps has often been captured manually (on digitising tables etc.) and 
the encoding of the semantic content has been seen as an extension of this. However, the successful automation of 
raster-vector conversion plus the large quantity of new and archived graphical data available on paper makes the 
automation of feature extraction desirable. 
Feature extraction and object recognition are large research areas in the field of image processing and computer vision. 
Recognition is largely based on the matching of descriptions of shapes. Numerous shape description techniques have 
been developed in computer vision, such as, boundary chain coding, analysis of scalar features (dimension, area, 
number of corners etc), Fourier descriptors and moment invariants. These techniques are well understood when applied 
to images and have been developed to describe shapes irrespective of position, orientation and scale. They can also be 
easily applied to vector graphical shapes. 
A description of the above four methods for shape recognition and their application for classifying objects on large- 
scale maps is described here. Unlike many applications where the shape categories are very specific (for example 
identifying a particular aircraft type in a scene), the problem requires the classification of a particular shape into a 
general class of similar object shapes, for example, building, road or stream. A comparison is made of the effectiveness 
of these techniques in recognising features on large-scale topographic maps and plans. 
  
480 International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B3. Amsterdam 2000.