الاستخلاص الآلي للنظم الأرضية باستخدام البيانات عالية الدقة لمرئيات اللاندسات و الأستر : دراسة حالة منطقة أبو رديس، جنوب غربي شبه جزيرة سيناء

Abstract EN

There is a need to automate terrain feature mapping so that to make the process more objective and less time consuming by using proper feature extraction techniques.

The objective of this study was the use of object-oriented image analysis methods for the automatic extraction of land systems.

The study area was located in Abu Rudeis area, South Western of Sinai Peninsula, Egypt.

The data used included a Landsat ETM+ image (bands1,2,3) in blue, green and red spectral ranges, with 30 meter spatial resolution, an ASTER L1 satellite image (bands1,2,3) with 15 meter spatial resolution and DEM with 15 meter spatial resolution.

A multiresolution segmentation technique was then developed, A Multispectral Image Segmentation Approach for Objectbased Image Classification of High Resolution Satellite Imagery delivering object primitives at different hierarchal levels.

Three physiographic feature types (Suez gulf water, alluvial fans of Baba, Sidri, and Wadi Feiran drainage basins and mountainous areas) were extracted.

The method gave good results in extracting land systems.

The results which obtained from traditional classification techniques and suggestion segmentation technique were compared on the level of both accuracy assessments of them, Image segmentation technique has been recognized as an essential process that performs an object-based rather than a pixel-based classification of highresolution satellite imagery.