Overview
MIRAGE is a Python-based framework designed for the signature-oriented extraction of geospatial evidence from multisource datasets, including Digital Elevation Models (DEMs), radiometric rasters, and geophysical grids.
It replaces the earlier PyLINE concept by shifting the focus from automated detection toward expert-led morphological identification.
Core Philosophy: The Signature Approach
Unlike black-box lineament extractors, MIRAGE operates through two distinct interpretative lenses to ensure that data extraction does not betray the nature of the terrain.
- The Geological Lens: Prioritizes patterns consistent with natural dynamics, such as regional tectonic orientations, structural continuity, and multifractal scaling behavior.
- The Archaeological Lens: Optimized for anthropogenic signatures, identifying discrete geometries, localized micro-relief, and non-linear patterns, including circular or pipe-like traces that are often lost in regional filtering.
Technical Capabilities
- Multisource Integration: Processes contrast-enhanced rasters, including RRIM, false color, and radiometry, to reveal signatures hidden to the naked eye.
- Advanced Extraction Logic: Employs multiscale filtering and edge detection structured by expert-defined parameters, ensuring every result is reproducible and defensible.
- Signature-Driven Workflow: Requires the declaration of an interpretative objective before execution, moving away from blind tracing toward rigorous evidence extraction.
- GIS-Ready Output: Generates high-fidelity vector data, including Shapefiles, designed for downstream fractal, multifractal, and geostatistical analysis.
Why MIRAGE?
In applied geophysics and archaeology, detecting an anomaly is easy; interpreting it is the challenge. MIRAGE treats ambiguity as a property of the data to be managed, not a flaw to be ignored.
It transforms remote sensing into a transparent, structured practice of morphological argument.
MIRAGE does not promise to see magic. It promises to see better.