HAHSI (Hexagonal Array for HyperSpectral Imaging)
Overview
Retrieving the reflectance spectrum from objects is an essential task for many classification and detection problems, since many materials and processes have a unique spectral behavior. In many cases, it is highly desirable to capture hyperspectral images due to the high spectral flexibility. Often, it is even necessary to capture hyperspectral videos or at least to be able to record a hyperspectral image at once, also called snapshot hyperspectral imaging, to avoid spectral smearing. For this task, a high-resolution snapshot hyperspectral camera array using a hexagonal shape is introduced. The Hexagonal Array for HyperSpectral Imaging (HAHSI) uses off-the-shelf hardware, which enables high flexibility regarding employed cameras, lenses, and filters. Hence, the spectral range can be easily varied by mounting a different set of filters. Moreover, the concept of using off-the-shelf hardware enables low prices in comparison to other approaches with highly specialized hardware. Since classical industrial cameras are used in this hyperspectral camera array, the spatial and temporal resolution is very high, while recording 37 hyperspectral channels in the range from 400 to 760 nm in 10 nm steps. As the cameras are at different spatial positions, a registration process is required for near-field imaging, which maps the peripheral camera views to the center view. This combination is used to provide a real-world high-resolution hyperspectral video database with ten scenes.
Database
The HAHSI database provides ten scenes recorded from 400 nm to 760 nm in 10 nm steps, resulting in 37 hyperspectral channels. The following table provides details about these scenes:
| Name | Resolution | Frame Rate | Frames | Exposure | Near-field | Far-field |
| Cars | 1600 x 1100 | 30 FPS | 31 | 5 ms | Y | |
| Cola Mix | 600 x 400 | 170 FPS | 3351 | 5 ms | Y | |
| Lab Pan | 1600 x 1200 | 30 FPS | 200 | 5 ms | Y | |
| Tree | 800 x 1300 | 50 FPS | 925 | 5 ms | Y | |
| Outdoor Pan 1 | 1124 x 924 | 23 FPS | 233 | 10 ms | Y | |
| Outdoor Pan 2 | 1124 x 924 | 23 FPS | 286 | 10 ms | Y | |
| Outdoor Pan 3 | 1124 x 924 | 23 FPS | 316 | 10 ms | Y | |
| Surveillance | 1124 x 924 | 23 FPS | 1000 | 4 ms | Y | |
| Run | 1132 x 928 | 22 FPS | 380 | 1 ms | Y | |
| Campus | 1132 x 928 | 22 FPS | 900 | 1 ms | Y |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
| Cars | Cola Mix | Lab Pan | Tree | Outdoor Pan 1 | Outdoor Pan 2 | Outdoor Pan 3 | Surveillance | Run | Campus |
Source Code
The GitHub repository provides a hyperspectral video viewer, which depicts the hyperspectral video as well as the corresponding RGB video rendered from the hyperspectral channels. The GitHub repository of another paper provides the source code required for registration.
Publication
If you use the dataset or source code for your research, you should cite the following paper:
Frank Sippel, Jürgen Seiler, André Kaup
High-resolution hyperspectral video imaging using a hexagonal camera array
Journal of the Optical Society of America A-Optics Image Science and Vision, vol. 41, num. 12, Dec 2024, pp. 2303-2315
DOI: 10.1364/JOSAA.536572
arxiv: https://arxiv.org/abs/2407.09038
License
The database and source are licensed using the BSD-3-Clause license.