Transparent minerals (such as quartz, calcite, muscovite, fluorite, and diopside) exhibit high transmittance and low reflectance across the visible spectrum. Consequently, their conventional diffuse reflectance signatures are weak, making traditional hyperspectral imaging insufficient for robust classification. However, transparent minerals typically display distinct crystalline anisotropy—variations in refractive index, absorption coefficients, and reflectance properties along different crystallographic axes that become significantly amplified under polarized light. Hyperspectral polarization imaging integrates linear polarizers or liquid crystal tunable polarizers into the optical path to sequentially acquire spectral datacubes at multiple polarization angles (e.g., 0°, 45°, 90°, and 135°), extracting vital parameters such as the Stokes vector components (I, Q, U), Degree of Linear Polarization (DoLP), and Angle of Polarization (AoP). Depending on their crystal structures, optical signs (uniaxial/biaxial, positive/negative), and birefringence, different transparent minerals manifest highly discriminative polarization-spectral response profiles, establishing a new physical dimension for mineral identification. For thin-section mineral identification, traditional petrographic microscopy requires complex sample preparation to grind specimens to a standardized thickness of 0.03 mm for the observation of interference colors and extinction behaviors under cross-polarized light—a process that is highly dependent on operator experience. In contrast, hyperspectral polarization imaging can scan hand specimens or cut core surfaces with minimal preparation or basic polishing. By computing variations in reflectance spectra under distinct polarization states and mapping DoLP distributions, it automatically classifies mineral species and crystal orientations. The HG-HyperLab laboratory hyperspectral imaging system, engineered by Hagorun Technology Limited, features an optional polarization lens assembly to achieve simultaneous acquisition of hyperspectral and polarization data streams, serving as an integrated experimental workspace for transparent mineral analysis. Regarding fluid inclusion and microstructural micro-prospecting, internal fluid-gas inclusions, growth zoning, and twin boundaries exhibit distinct polarization anomalies driven by localized refractive index contrasts. These show up as characteristic diagnostic patterns in hyperspectral polarization imagery, assisting in modeling mineral genesis and ore-fluid evolution paths.

In gemology and jade authentication, hyperspectral polarization imaging distinguishes between transparent gemstone species with overlapping visual phenotypes. For instance, colorless sapphire, spinel, and zircon display minimal contrasts in conventional visible-range reflectance spectra. Under polarized illumination, however, their varied refractive indices and optical properties yield distinct DoLP and AoP spatial profiles. Spinel, as an isometric (isotropic) phase, exhibits no polarization effect. Colorless sapphire, being anisotropic, displays systematic periodic fluctuations in reflectance across varying polarization angles, while zircon exhibits strong, characteristic birefringence signatures. Extracting polarization response curves and DoLP histograms from specific regions of interest establishes classification thresholds for rapid, non-destructive gemstone discrimination, avoiding destructive testing. In the speciation of mica group minerals (e.g., muscovite, biotite, phlogopite, lepidolite), traditional classification relies on optical configurations, density measurements, and destructive chemical assays, making for a cumbersome workflow. Hyperspectral polarization imaging isolates shifts in polarization-spectral features caused by compositional substitutions within the mica crystal lattice: muscovite presents characteristic Al-OH overtones in the short-wave infrared window, biotite displays lower polarized reflectance in the visible band due to elevated iron content, and the magnesium content of phlogopite distinctly shifts its polarization response profile. Integrating DoLP spectral shapes with specific band AoP indices trains classification models for the automated sorting of mica phases. The HG-HyperLab hyperspectral imaging system from Hagorun Technology Limited, paired with polarization accessories and custom analytics software, supports full polarization-spectral data preprocessing, Stokes parameter calculation, and automated classification model construction, delivering a reliable workflow tool for geological laboratories. For carbonate mineral discrimination (such as distinguishing calcite, dolomite, and magnesite), the phases share analogous crystal habits and overlapping reflectance spectra but diverge in their polarization properties due to varying cation species. Computing the ratios of surface reflectance under distinct polarization states alongside DoLP spectral trends allows for the clear differentiation of calcite from dolomite. This methodology holds valuable application potential for studying sedimentary diagenesis and evaluating carbonate reservoir architectures.

The crystallographic orientation of transparent minerals dictates their bulk physical properties and industrial value. By analyzing the polarization response profiles of mineral surfaces across varying orientations, hyperspectral polarization imaging can non-destructively determine preferred crystal orientations. Under cross-polarized setups, extinction occurs when the mineral's optic axis aligns at specific orientations relative to the polarization vector. Rotating the specimen or shifting the polarization angle while recording reflectance variations via polar plots allows analysts to map the mineral's optical sign and optic axis configuration. This methodology is highly effective for fabric and texture analysis of aligned micas or amphiboles in metamorphic rocks, offering insight into structural deformation histories. In gemstone rough orientation and pre-cutting optimization, hyperspectral polarization imaging assists in planning the optimal cutting vectors for raw crystals. For gemstones displaying strong pleochroism (such as sapphire, tourmaline, and andalusite), differential light absorption along separate crystallographic axes directly alters the final gem's color saturation and brilliance. Scanning with polarized hyperspectral optics quantifies transmission and reflection polarization behaviors along distinct structural axes, providing empirical data to guide cutting operations and minimize material loss. Furthermore, hyperspectral polarization imaging has strong potential for distinguishing synthetic crystals from natural gems. Synthetic phases typically feature highly uniform polarization responses and negligible internal stress fields. Natural crystals, conversely, exhibit complex irregular internal strain distributions from their variable growth environments, manifesting as diagnostic patchy anomalies in DoLP maps. The HG-HyperLab hyperspectral imaging system from Hagorun Technology Limited supports seamless switching between macro and microscopic imaging scales, accommodating samples ranging from centimeter-scale rock slabs to millimeter-scale gemstone grains, providing an adaptable polarization-spectral configuration for research across geosciences, gemology, and materials engineering.

Interested in Advancing Hyperspectral Imaging Applications?

Our engineering group delivers advanced technical consulting and integrated solutions

Hagorun Technology Limited  |  Focus · Dedication · Exploration · Vision