Satellite Image Automatic Mapper™  (SIAM™ ) - A Turnkey Software Executable for Automatic Near Real-Time Multi-Sensor Multi-Resolution Spectral Rule-Based Preliminary Classification of Spaceborne Multi-Spectral Images

Andrea      Baraldi

Abstract

It is an unequivocal fact that to date the automatic or semi-automatic transformation of huge amounts of multisource multi-resolution remote sensing (RS) imagery into useful information (such as biophysical variables, categorical maps, etc.) still remains far more problematic than might be reasonably expected. To invert this trend, a novel two-stage stratified hierarchical hybrid remote sensing image understanding system (RS-IUS) was presented in recent literature to encompass the four levels of analysis of an information processing device, namely, computational theory (system architecture), knowledge/information representation, algorithm design and implementation. The proposed original RS-IUS architecture comprises: (i) a first-stage pixel-based application-independent top-down (deductive, physical model-driven, prior knowledge-based) preliminary classifier and (ii) a second-stage battery of stratified hierarchical context-sensitive application-dependent modules for class-specific feature extraction and classification. The proposed implementation of the prior knowledge-based preliminary classification first stage of a two-stage stratified hierarchical hybrid RS-IUS consists of the original Satellite Image Automatic Mapper™ (SIAM™ , University of Maryland Invention Disclosure No. IS-2010-103, patent pending, © Andrea Baraldi & University of Maryland). SIAM™ is an operational automatic (turnkey, good-to-go, press-and-go) software button (executable) for unsupervised near real-time per-pixel multi-source multiresolution application-independent spectral rule-based decision-tree classification of spaceborne multi-spectral imagery. The goal of this patent review is to highlight the several degrees of novelty and operational advantages of the proposed two-stage hybrid RS-IUS employing SIAM™ as its preliminary classification first stage in comparison with alternative approaches, such as two-stage object-based RS-IUSs which, in spite of their lack of consensus and research, have recently gained widespread popularity in both scientific and commercial RS image applications.

Keywords: Decision-tree classifier, deductive learning, image classification, inductive learning, prior knowledge, radiometric calibration, remote sensing, Satellite Image, spaceborne mission, RS-IUS architectures