Hyperspectral imaging (HSI) has been used for a couple of decades in applications such as satellite imaging, air reconnaissance and other not overly price sensitive markets. Still, there is no clear definition of hyperspectral imaging. Sometimes techniques that produce 2D images with more than the typical three RGB colors or spectral channels -for example by inclusion of a near-infrared channel -are already called hyperspectral. Mostly though, this is not considered sufficient. Typically, even ten spectral channels are still to be called multi-spectral rather than hyperspectral.In the following, we require that certain criteria are fulfilled for an imaging technique to be hyperspectral: ■ ■ For every pixel in the image, we measure the spectrum of the incident light or radiation.■ ■ The measured spectrum is continuous and not discretized to a limited number of channels.■ ■ The spectrum covers more than one sub-wavelength range, for example UVA, visible and near-infrared or NIR and SWIR. The advent of alternative approaches makes HSI attractive for volume markets or even consumer products, for example cancer detection, precision farming with UAV or directly at the plant, food testing in supermarkets and many more.Alternative approaches comprise sensors that are coated at wafer-level with fixed wavelength bandpass filters, e.g. Pixelteq or imec. Common are also thin film coatings on glass substrates that can be patterned during deposition (in situ), or by using a photolithographic process over the coating to block the addition or subtraction of materials deposited on the substrate surface, e.g. Materion. These micro-patterning techniques allow either filters that have a staircase of different center wavelengths in one direction, suited for the so-called push broom technique, or 2D mosaics, suited for the so-called snapshot technique.However, according to the above criteria, strictly speaking these solutions do not provide hyperspectral capability but are inherently multispectral due to their discrete changes in center wavelength. Typically, between 10 and 100 different wavelengths or channels are offered. A truly hyperspectral sensor offers a continuous change in center wavelength and as such a virtually unlimited number of channels.Delta Optical Thin Film follows a different approach to filters for hyperspectral imaging. Based on our extensive experience with linear variable filters, we develop and manufacture custom linear variable bandpass filters (LVBPF) for mid-size and full-frame CCD/CMOS sensors (e.g. 25 mm × 25 mm or 24 mm × 36 mm). These filters offer very high transmission levels and are fully blocked in the light-sensitive wavelength range of silicon-based detectors (200 nm to 1150 nm or higher). The combination of LVBPFs with silicon detectors allows the design of very compact, robust and affordable HSI detectors that offer sev- Delta Optical Thin Film A/S is the leading supplier of advanced, high performance linear variable filters commonly used in a variety of biomedical imaging applications including...

Linear Variable Filters recently reached quality levels that make them applicable for demanding biomedical and hyperspectral imaging applications that often rely on fluorescence techniques. LVF offer tuneability over a wide wavelength range with excellent blocking.

No abstract

Hyperspectral and multispectral imaging have been used for a couple of decades in applications such as satellite imaging, air reconnaissance, and other not overly price sensitive markets. The advent of alternative approaches makes spectral imaging attractive for volume and consumer markets, for example cancer detection, precision farming with unmanned aerial vehicles (UAV) or directly at the plant, or food testing in supermarkets. Alternative approaches comprise wafer-level coated sensors with fixed wavelength bandpass filters. Thin film coatings on glass substrates that can be patterned during deposition (in situ), or by using a photolithographic process over the coating to block the addition or subtraction of materials deposited on the substrate surface are also common. These micro-patterning techniques allow filters in a 2D mosaic struc-tors that offer several advantages and benefits over conventional approaches:  huge aperture compared with grating and prism,  higher transmission than grating and prism,  short measurement time,  high suppression of stray light,  excellent signal to background ratio,  3D and snapshot capability. Fig. 1 shows the transmission characteristics of a CVBPF that covers a center wavelength range in VIS/NIR with a bandwidth of approximately 2 % of its center wavelength. In a wide wavelength range, the transmission is higher than 90 %. But even more important than the peak transmission all undesired radiation from 200 to 1150 nm is suppressed better than OD4. ture (suited for the snapshot technique, i.e. acquisition of the hyperspectral data cube with only one camera exposure). A smart combinationA novel snapshot multispectral camera with high spectral performance and high spatial resolution was achieved by combining a micro lens array imaging system by German Fraunhofer IOF and a continuously variable bandpass filter by Danish Delta Optical Thin Film. The continuously variable bandpass filters (CVBPF) developed and manufactured by Delta Optical Thin Film offer high transmission and are fully blocked in the light-sensitive wavelength range of silicon-based detectors. The combination of CVBPFs with silicon detectors (Fig. 2) allows the design of compact, robust, and affordable spectral imaging detec-A multispectral imaging concept based on a multi-aperture system approach using a customized microlens array combined with a slanted continuously variable bandpass filter and a silicon-based image sensor helps to overcome the restrictions of scanning techniques or wafer-level coated detectors.Snapshot multispectral camera demonstrator (Source all images: the authors)

No abstract