Spectrum Investigation TRIZ Cases
- Systems and methods for differentiable programming for hyperspectral unmixing
Differentiable programming simulates endmember variability and iteratively estimates material abundances with physically plausible results.
- Optical Spectrum Peak Detection with Zone-Trained Neural Networks
A zone-trained neural network detects more optical spectrum peaks while limiting crosstalk sensitivity without adding detectors.
- Imaging Spectral Sensor Fusion for Localized White Balance Correction
Fabry-Perot spectral filters combine with higher-resolution imaging to correct illumination locally across complex scenes.
- Multispectral Imaging Device for Single-Camera Optical Inspection
A single camera combines N-band filtering, varied illumination, and matrix conversion to inspect different optical targets at one angle.
- Bragg Resonance Spectral Filter for Miniaturized Image Sensors
Shared Bragg reflective layers and cavity resonance combine wavelength filtering in a compact structure for chip-integrated image sensors.
- Photothermal Imaging With Thermal-Lens Detection Beyond Mid-IR Diffraction
A mid-infrared pump and visible probe detect thermal lenses for fast, sensitive imaging without cooled mid-IR detectors.
- Optical Filter Coatings for Low-Cost Precision Measurement
Coat, divide, test, and select filter portions to simplify optical systems while preserving measurement accuracy and wavelength resolution.
- Diffractive Imaging Device for Single-Shot Hyperspectral Capture
This imaging device uses diffraction and segmented sensor regions to capture multispectral data with less system complexity and computation.
- Color Batch Correction with Spectral Coordinates and Machine Learning
Machine learning converts panel spectra into L*a*b* coordinates and predicts coating adjustments, reducing iterations and material waste.
- Spectral Sensor pH and SmO2 Detection for Cardiac Arrest Timing
Measure muscle pH and SmO2 through skin to estimate down time and assess treatment during cardiac arrest.
- CMOS Image Sensor With Pixel-Specific Filters for AI Spectral Data
Pixel-specific filters expand spectral diversity in CMOS sensors while preserving light sensitivity for richer AI imaging inputs.
- Multi-Layered Cell Mass Quality Evaluation Using Spectral Light Variation
Measure 300–2,000 nm light intensity variation across cell masses to assess quality without lengthy time-series imaging.