Hyperspectral imaging is a different way of collecting photographic information. It is much more powerful than traditional photography, and provides significantly more information about a scene than can be seen with the naked eye. Hidden details can be found, and material properties can be discovered. Hyperspectral data can provide information to enable and accelerate the automation of tasks such as in the detection of crop disease, forensic applications and differentiating between different materials.
Hyperspectral cameras are now available at a fraction of the cost than in the past. They offer a user the ability to capture more information about a scene than can be done with a standard RGB camera. Captured images may have tens or hundreds of bands depending on the hardware used, with each channel of the resultant image describing captured light for a given wavelength. The resultant data is an 'image cube' (two dimensions being spatial and the third being spectral). Further processing of this image cube allows for accurate scene analysis in a range of applications far beyond that of an ordinary RGB image, for example, the 'spectral signatures' for each pixel can be used to discover the material composition of a scene.
Scyllarus hyperspectral imaging software can be used to assist in processing hyperspectral images. The benefits associated with using this software includes increases in accuracy of analysis (for example, by removing effects due to lighting and surface reflection on objects), and the reduction in constraints at the time of capture (for example, aerial imaging could be conducted at a wider variety of times of the day due to lighting compensation and ground-based use can be achieved more practically). An important feature of the software is the ability to perform tasks such as material indexing without the need to calibrate for lighting.
Features of this technology include:
- camera agnostic, compatible with all raw hyperspectral data
- load and save hyperspectral image data in a number of formats
- automatic illuminant recovery – the only input is the image
- automatic reflectance recovery – no calibration required
- automatic spectra-based material clustering/segmentation, fast estimation method available
- principal component analysis and spectral angle mapping algorithms
- load and save spectral libraries and perform spectral unmixing
- NURBS (Non-Uniform Rational B-Spline) encoding/decoding of spectral signatures and images
- fast and low memory overhead hyperspectral image manipulation features: crop, resize, remove bands, align bands
- streamlined pipeline processing model – adding an image and processing for results takes only a few lines of code
- algorithms optimised for multi-processor machines
- accelerated by Armadillo, BLAS, LAPACK, OpenCV, OpenMP and Boost.
In environments where traditional RGB imaging is now used, hyperspectral imaging can provide more flexible and better results. Applications such as food inspection, (fruit sorting or inspecting laboratory samples for bacteria or micro-organisms), agriculture (early detection of diseases in plants before they are visible, automated crop identification and health monitoring), and security (face detection allowing more flexible lighting and the ability to more quickly identify skin variants, accurate object tracking, etc) can all be improved through the use of hyperspectral data.
The Scyllarus C++ API can be used by developers to integrate hyperspectral processing into an application, or to build an application that makes use of the features of Scyllarus. Scyven, the 'Scyllarus Visualisation Environment' is built upon the Scyllarus C++ API; to see the features of the API in action, check it out at Scyven - Hyperspectral Visualisation Tool. The Scyllarus C++ API includes many powerful image processing features and tools targeted at hyperspectral imaging applications. With an easy to use modular design, it's simple to integrate the Scyllarus API into your project.
CSIRO has a number of granted and pending patent families that cover various aspects of the Scyllarus Hyperspectral Technology.
The team consists of a number of researchers and engineers, with many years of experience in the fields of imaging spectroscopy and multi/hyperspectral imaging.