Larisa Poryvkina

Total luminescent spectroscopy for remote laser diagnostics of natural water conditions

In undertaking the rapid diagnostics of water conditions (for organic impurities, eutrophication etc.), it is expedient to confine the effort to establishing the statistically normal state and departures from it rather than to seek the full chemical composition of the medium. What others have termed spectral signatures obtained by such methods as total luminescent spectroscopy can serve as representative indicators of both background conditions and those deviating from the normal. In this paper we show how, with the aid of laboratory catalogs of spectral signatures, it is possible to detect oil pollution at levels substantially below those of slicks (of a few microliters per liter) and to identify groups of pollutants (by remote sensing). And we describe a spectroscopic lidar we developed for marine studies.

Spectral pattern recognition of controlled substances in street samples using artificial neural network system

The NarTest fluorescent technique is aimed at the detection of analyte of interest in street samples by recognition of its specific spectral patterns in 3-dimentional Spectral Fluorescent Signatures (SFS) measured with NTX2000 analyzer without chromatographic or other separation of controlled substances from a mixture with cutting agents. The illicit drugs have their own characteristic SFS features which can be used for detection and identification of narcotics, however typical street sample consists of a mixture with cutting agents: adulterants and diluents. Many of them interfere the spectral shape of SFS. The expert system based on Artificial Neural Networks (ANNs) has been developed and applied for such pattern recognition in SFS of street samples of illicit drugs.

Standoff detection: classification of biological aerosols using laser induced fluorescence (LIF) technique

The challenges of detecting hazardous biological materials are manifold: Such material has to be discriminated from other substances in various natural surroundings. The detection sensitivity should be extremely high. As living material may reproduce itself, already one single bacterium may represent a high risk. Of course, identification should be quite fast with a low false alarm rate. Up to now, there is no single technique to solve this problem. Point sensors may collect material and identify it, but the problems of fast identification and especially of appropriate positioning of local collectors are sophisticated. On the other hand, laser based standoff detection may instantaneously provide the information of some accidental spillage of material by detecting the generated thin cloud. LIF technique may classify but hardly identify the substance. A solution can be the use of LIF technique in a first step to collect primary data and – if necessary- followed by utilizing these data for an optimized positioning of point sensors. We perform studies on an open air laser test range at distances between 20 and 135 m applying LIF technique to detect and classify aerosols. In order to employ LIF capability, we use a laser source emitting two wavelengths alternatively, 280 and 355 nm, respectively. Moreover, the time dependence of fluorescence spectra is recorded by a gated intensified CCD camera. Signal processing is performed by dedicated software for spectral pattern recognition. The direct comparison of all results leads to a basic classification of the various compounds.

Standoff detection and classification procedure for bioorganic compounds by hyperspectral laser-induced fluorescence

The high and still increasing number of attacks by hazardous bioorganic materials makes enormous demands on their detection. A very high detection sensitivity and differentiability are essential, as well as a rapid identification with low false alarm rates. One single technology can hardly achieve this. Point sensors can collect and identify materials, but finding an appropriate position is time consuming and involves several risks. Laser based standoff detection, however, can immediately provide information on propagation and compound type of a released hazardous material. The coupling of both methods may illustrate a solution to optimize the acquisition and detection of hazardous substances. At DLR Lampoldshausen, bioorganic substances are measured, based on laser induced fluorescence (LIF), and subsequently classified. In this work, a procedure is presented, which utilizes lots of information (time-dependent spectral data, local information) and predicts the presence of hazardous substances by statistical data analysis. For that purpose, studies are carried out on a free transmission range at a distance of 22m at two different excitation wavelengths alternating between 280nm and 355 nm. Time-dependent fluorescence spectra are recorded by a gated intensified CCD camera (iCCD). An automated signal processing allows fast and deterministic data collection and a direct subsequent classification of the detected substances. The variation of the substance parameters (physical state, concentration) is included within this method.