Hyperspectral Object Identification via Segmented Equation Solving

The patent segments the object into multiple parts (e.g., different sections of a launch vehicle) and creates separate equations for each part. By dividing the object model into discrete segments with unique temperature and emissivity characteristics, the system can resolve and identify individual parts even when they appear spatially unresolved in the sensor data, thereby maintaining identification capability at long ranges.

The patent transitions from spatial dimension analysis to spectral dimension analysis by utilizing hyperspectral data across multiple bands. Instead of relying on spatial resolution to distinguish object parts, the system uses spectral signatures (emissivity and temperature variations across different wavelength bands) to differentiate between parts, effectively moving the discrimination task to a different dimension where resolution is preserved.

The patent creates a universal model that simultaneously handles multiple types of information (spatial, thermal, and emissivity) through a unified system of equations. The model integrates radiometric measurements with physical properties (emissivity and temperature) of different object parts, allowing a single system to extract diverse information types that would otherwise require separate measurement systems.

The patent changes the measurement parameters from purely spatial to include thermal and emissivity parameters. By incorporating temperature and emissivity as additional dimensions in the analysis model, the system recovers detailed thermal and emissivity information that would be lost in traditional imaging, transforming the data extraction process to capture multiple physical properties simultaneously.

The patent segments the complex hyperspectral dataset into manageable components by creating separate equations for each object part. This segmentation allows the processing system to handle one part at a time rather than attempting to process all spectral bands for all parts simultaneously, reducing computational complexity while preserving complete information about each segment.

The patent applies local quality by assigning unique temperature and emissivity characteristics to each object part rather than treating the entire object uniformly. This localized approach allows the system to process and identify each part with its specific properties independently, simplifying the overall processing by breaking down the global problem into localized sub-problems that can be solved more efficiently.