1：Experimental Design and Method Selection:

The experiment involved collecting hyperspectral data over the RIT campus using an aerial sensor, with a focus on target detection and atmospheric compensation. Methods included the Empirical Line Method (ELM) for atmospheric compensation and radiative transfer modeling with MODTRAN for radiometric fidelity assessment.

2：Sample Selection and Data Sources:

The dataset includes 90 painted wood block targets (45 green, 45 yellow) deployed over a 380m by 260m area, with background pixels up to 8 million. Data sources include hyperspectral imagery, LiDAR, high-resolution RGB images, and field spectral measurements.

3：List of Experimental Equipment and Materials:

Equipment includes a silicon-based Headwall imaging spectrometer (400-1000 nm), LiDAR, an 80-megapixel RGB camera, and field spectrometers (e.g., ASD, SVC). Materials include painted wood blocks and large calibration panels (red, blue, green, brown, white, black).

4：Experimental Procedures and Operational Workflow:

Data collection occurred on August 1, 2017, with flight lines at 5000 feet altitude. Targets were placed on various backgrounds, and spectral measurements were taken. Radiometric fidelity was checked using MODTRAN, and atmospheric compensation was applied using ELM with calibration panels.

5：Data Analysis Methods:

Analysis involved comparing predicted and measured spectral radiance, applying gain and bias corrections for reflectance retrieval, and using spectral smoothing to reduce noise.