You are using an outdated browser that does not fully support the intranda viewer.
As a result, some pages may not be displayed correctly.

We recommend you use one of the following browsers:

Full text

Fusion of sensor data, knowledge sources and algorithms for extraction and classification of topographic objects
Baltsavias, Emmanuel P.

International Archives of Photogrammetry and Remote Sensing, Vol. 32, Part 7-4-3 W6, Valladolid, Spain, 3-4 June, 1999
The Joint ISPRS / EARSeL Workshop “Fusion of sensor data, knowledge sources and algorithms for extrac
tion and classification of topographic objects”, whose proceedings constitute this publication, brought together
specialists in Remote Sensing, Photogrammetry, Image Analysis and GIS to discuss the current state of
(semi-)automatic procedures for acquiring geometric and semantic object information from images and digital
databases. It dealt mainly with extraction and classification of objects like landcover, forest areas, urban settle
ments, and specific objects like buildings or roads. Various types of spaceborne and airborne sensors were
used (panchromatic, multispectral, hyperspectral and microwave imagery, airborne laser scanners). The goal
of the Workshop was to unify activities of working groups with similar and partly overlapping terms of refer
ence, coordinate actions between ISPRS and EARSeL (a regional ISPRS member) and permit in-depth dis
cussions on common significant and hot topics like object models, and analysis methods and strategies with
emphasis on the integration of multiple information, knowledge sources and algorithms. The topics referred to
basic research or advanced applications of Photogrammetry, Remote Sensing and Computer Vision for geo
database generation and update. The emphasis was on the development of methods and algorithms for data
processing in various application fields and less on data acquisition.
More specifically the workshop dealt with:
• Introduction, overview and concepts of image / data / information fusion and integration
• Tools and methods for multiresolution and multitemporal image fusion
• Exploitation of various object cues like DSMs, spectral information, texture and backscatter anisotropy, and
morphology for object extraction and recognition
• Use of GIS and map databases to help the object extraction and recognition process
• New classification methods
• Image to image and vector registration
• Fusion of DEMs from various sensors or with various resolutions
• Applications in forestry
Some of the characteristics of the workshop include:
• Cooperation of ISPRS Working Groups spanning three Commissions and one EARSeL Special Interest
• 27 papers from 11 countries presented in 8 Technical Sessions
• Mixture of academia (75%) and research institutes (20%), with little participation from public organisations
and private companies
• 68% of the authors have a remote sensing/photogrammetry background, 20% come from electrical engi
neering, physics and computer science, and 12% from planning, forestry, environmental sciences.
• Publication of the workshop proceedings in the ISPRS Archives Series
The submitted abstracts were evaluated by the scientific committee. The general interest that was expressed
makes us confident that the issues that are dealt with in this publication are relevant to a great number of
experts and users in the related fields. The papers are interesting, variable, coherent, and focussed on the
workshop topics in spite of the different backgrounds of the authors.
From this position we would like to thank all authors for their contribution to the success of the Workshop.