Automatic Archaeological Feature Extraction From Satellite VHR Images by Mathematical Morphology Functions

Paper number

IAC-08.B1.4.11

Author

Dr. munzer jahjah, Italy

Coauthor

Mr. GIANCARLO SANTILLI, Italy

Coauthor

Prof. Carlo Ulivieri, University of Rome "La Sapienza", Italy

Year

2008

Abstract

Archaeological applications need a methodological approach on a variable scale able to satisfy the intra-site (excavation) and the inter-site (survey, environmental research). The increased availability of high resolution and micro-scale data has substantially favoured archaeological applications and the consequent use of GIS platforms for Reconstruction of archaeological landscapes based on remotely sensed data
Feature extraction of multispectral remotely sensing image is an important task before any further processing. High resolution remote sensing data, specially panchromatic, is an important input for the analysis of various types of image characteristics; It plays an important role in the visual systems for recognition and interpretation of given data.
The methods proposed rely on an object-oriented approach based on a theory for the analysis of spatial structures called Mathematical Morphology. The term "morphology" stems on the fact that it aims at analysing object shapes and forms. It is mathematical in the sense that the analysis is based on the set theory, integral geometry, and lattice algebra. Mathematical Morphology has proven to be a powerful image analysis technique; two-dimensional gray tone images are seen as three-dimensional sets by associating each image pixel with an elevation proportional to its intensity level. An object of known shape and size, called the structuring element, is then used to investigate the morphology of the input set. This is achieved by positioning the origin of the structuring element to every possible position of the space and testing, for each position, whether the structuring element either is included or has a nonempty intersection with the studied set. The shape and size of the structuring element must be selected according to the morphology of the searched image structures.
These techniques were applied to different archaeological sites in Turkmenistan (Nissa) and in Iraq (Babylon); A further change detection analysis were applied to the Babylon site using two HR images as a pre -post second golf war.
We had different results or outputs, taking in consideration that the operative scale of sensed data determines the final result of the elaboration and the output of the information quality, because each of them was sensitive to specific shapes in each input image, we had mapped linear and non-linear objects, updating archaeological cartography, automatic change detection analysis for the Babylon site. The discussion of these techniques has the objective to provide the archaeological team with new instruments for the orientation and the planning of a remote sensing application.

Abstract document

IAC-08.B1.4.11.pdf

Manuscript document

IAC-08.B1.4.11.pdf (🔒 authorized access only).

To get the manuscript, please contact IAF Secretariat.