| _version_ | 1866901755939258368 |
|---|---|
| author | Nitu, Andreea Racoviteanu, Andrei Florea, Corneliu Ivanovici, Mihai |
| author_facet | Nitu, Andreea Racoviteanu, Andrei Florea, Corneliu Ivanovici, Mihai |
| contents | <p>Recognizing and classifying crops using satellite-based remote sensing is important for reliable agricultural monitoring and precision farming. This study evaluates the performance of two imaging technologies—multispectral and hyperspectral—in distinguishing crop types under varying conditions. The Mahalanobis distance is employed as a separability metric and applied to annotated data. We compare hyperspectral imagery from PRISMA with multispectral data from Sentinel-2, focusing on a test region across different months of the year. The results demonstrate that PRISMA’s hyperspectral data generally achieves superior separability compared to Sentinel-2. However, factors such as vegetation canopy and spatial resolution can influence the results. These findings highlight the advantages of hyperspectral, but also the situations in which multi-spectral can be a reliable substitute.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_16534958 |
| institution | Zenodo |
| language | |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Multispectral versus Hyperspectral Imaging for Crop Separability Nitu, Andreea Racoviteanu, Andrei Florea, Corneliu Ivanovici, Mihai <p>Recognizing and classifying crops using satellite-based remote sensing is important for reliable agricultural monitoring and precision farming. This study evaluates the performance of two imaging technologies—multispectral and hyperspectral—in distinguishing crop types under varying conditions. The Mahalanobis distance is employed as a separability metric and applied to annotated data. We compare hyperspectral imagery from PRISMA with multispectral data from Sentinel-2, focusing on a test region across different months of the year. The results demonstrate that PRISMA’s hyperspectral data generally achieves superior separability compared to Sentinel-2. However, factors such as vegetation canopy and spatial resolution can influence the results. These findings highlight the advantages of hyperspectral, but also the situations in which multi-spectral can be a reliable substitute.</p> |
| title | Multispectral versus Hyperspectral Imaging for Crop Separability |
| url | https://doi.org/10.5281/zenodo.16534958 |