Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

A chlorophyll fluorescence imaging system based on a filtered xenon lamp, providing actinic light in UV and violet bands, and on a high resolution camera equipped with a 690nm (FWHM=10nm) passband filter, for single band measurements, and with a four bands beam splitter with pass-band filters (450nm, 550nm, 690nm, 740nm, all with a FWHM=10nm), for multispectral measurements, was implemented and applied on wheat plants inoculated with different fungal infections, with the aim of investigating the potential of such a technique for detecting plants disease symptoms.

In steady state fluorescence images of attached leaves acquired at 690nm in laboratory conditions, the symptoms appear as highly emitting spots at sub-millimetric or millimetric scale which, with the progress of the disease, develop in larger, low emitting lesions surrounded by high intensity halo. Even if the changes in emission pattern are limited to the neighborhoods of the infection point; this technique allowed to detect disease presence before visible symptoms appear.

Kinetic fluorescence imaging performed by acquiring a sequence of images at 690nm during an actinic illumination period of several minutes, allowed to find differences between diseased and healthy areas, even at very early stages, both in terms of intensity and time-dependence of emission. Nevertheless, the excitation/sensing period of several minutes on which this technique is based, limits practical field applications on moving vehicles.

Multispectral fluorescence imaging in field conditions resulted unsuccessful during day-time measurements due to plants saturation by long-exposure to direct sunlight and to the interference of the diffuse environmental illumination. On the contrary, night-time imaging confirmed the high potential of this technique for disease detection and quantifications. In particular, an image analysis algorithm based on the ratio of fluorescence images at 550nm and 690nm was implemented, allowing to discriminate plants lesions and to map the disease severity in experimental plots in agreement with visual inspection made by a pathologist.