The Remote Sensing Toolbox is a comprehensive MATLAB-based framework designed for radiometric calibration, multi-spectral image analysis, land‑cover classification, change detection, spatial statistics, and publication-ready visualization. It integrates advanced calibration pipelines, including linear regression, histogram-based normalization, and per-band alignment methods, supporting diverse sensors such as Sentinel‑2 and PlanetScope. The toolbox provides eight unsupervised and eleven supervised classification algorithms, a full suite of vegetation indices, raster algebra tools, and robust statistical testing, including RMSE, MAE, t‑tests, Wilcoxon, bootstrap confidence intervals, and Moran’s I spatial autocorrelation. Additional modules enable time‑series trend analysis, AOI masking, resampling, RGB composites, deep-learning segmentation previews, and automated reporting through high‑resolution exports and multi-page PDF generation. Its modular architecture (spanning single‑date workflows, batch temporal processing, classification toolboxes, diagnostic visualization, and geospatial export) supports end‑to‑end remote sensing analysis for research and applied geospatial workflows.
Remote Sensing Toolbox
Christian Massimiliano Baldin
;Vittorio CASELLA
2026-01-01
Abstract
The Remote Sensing Toolbox is a comprehensive MATLAB-based framework designed for radiometric calibration, multi-spectral image analysis, land‑cover classification, change detection, spatial statistics, and publication-ready visualization. It integrates advanced calibration pipelines, including linear regression, histogram-based normalization, and per-band alignment methods, supporting diverse sensors such as Sentinel‑2 and PlanetScope. The toolbox provides eight unsupervised and eleven supervised classification algorithms, a full suite of vegetation indices, raster algebra tools, and robust statistical testing, including RMSE, MAE, t‑tests, Wilcoxon, bootstrap confidence intervals, and Moran’s I spatial autocorrelation. Additional modules enable time‑series trend analysis, AOI masking, resampling, RGB composites, deep-learning segmentation previews, and automated reporting through high‑resolution exports and multi-page PDF generation. Its modular architecture (spanning single‑date workflows, batch temporal processing, classification toolboxes, diagnostic visualization, and geospatial export) supports end‑to‑end remote sensing analysis for research and applied geospatial workflows.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
