The global pesticide market has grown steadily since the 1940s, with the agricultural sector being the largest user of pesticides. The effect of pesticides on human health is manifested either through direct exposure to the material or indirect exposure to contaminated resources. Farmers and those dwelling near areas where pesticides are used may suffer from direct exposure, while the general population might be exposed indirectly, for example, by drinking contaminated water. Exposure to pesticides may cause a variety of symptoms, including headaches, dizziness, and vomiting, damage the nervous system, and even cause death. The risks involved in pesticide use include air pollution and soil and water contamination. The environmental implications of pesticide use include development of resistance among pests, a decline in biodiversity, interruption of the food chain, and disruption of the ecological balance. Pesticide use may also cause changes in physical parameters of the ecosystem. Effective activity of pesticides requires reaching proper leaf coverage. To prevent pest attacks due to insufficient leaf coverage, farmers tend to apply pesticides in excess. In view of the environmental and health implications of pesticide use, there is a clear need to limit pesticide application. Yet farmers lack the means to perform real-time in situ assessment of leaf coverage. Existing pesticide detection methods are complex, time-consuming, and unsuited to field application. Optical methods have the potential to provide quick assessments and can be used in situ. Several optical methods for detection of pesticides in general and on leaves in particular were developed. The findings indicated that the main problems in pesticide detection using fluorescence are the low autofluorescence of the pesticides and the nonreproducible spectral response of the leaves. These obstacles were solved by employing labeling agents. For example, rhodamine was suggested, mainly due to its excellent surface adhesion and its extremely high fluorescence quantum yield. The labeling agents were sprayed on leaves in the form of aerosols, thus creating a uniform layer of nanocrystals and microcrystals on the surface of the leaves. The effects of pesticides on the spectral characteristics of the labeling agents were examined using laser-induced fluorescence (LIF) spectroscopy. When pesticide droplets were applied to a pretreated leaf, two phenomena were observed. The first was a substantial fluorescence increase. The second was material-specific spectral shifting as a result of interaction between the labeling molecules and organic components in the pesticide droplet. It was possible to utilize these spectral shifts for quantification of the pesticide concentration in the droplet. These spectral shifts enabled detection of pesticides on plants, although they were not sufficient for providing quantitative information on the extent of pesticide coverage. To detect pesticide coverage, several imaging data techniques were applied, such as LIF scanning of the examined plant surface. This method revealed the droplet shape by scanning and recording the fluorescence intensity at many points on a grid. Since application of this method is expensive and time-consuming, a second technique was also developed: it requires only a UV source and a CCD camera and it enables direct imaging of the pesticides on plants. The data obtained included the droplet shape and its location on the plant. When pesticide identification was required, application of a special hyperspectral fluorescence imaging mehod was introduced. Fourier transform hyperspectral imaging analysis provided simultaneous full spectral resolution at each pixel, enabling identification of the pesticide and its mapping on the plant. In practice, test plants have to be pretreated with labeling material before pesticide application. The changes in the labeling compound fluorescence can then be used for detection of the pesticide on the plant and quantification of the overall coverage. Low-cost mapping of the pesticide microdroplets could be obtained using a CCD camera, while accurate information could be based on Fourier transform hyperspectral imaging. Since these methods provide immediate results, they may allow the farmer to estimate leaf coverage during pesticide application and adjust spraying accordingly.
