Programmed strategies for an early discovery of plant illnesses are indispensable for accuracy crop assurance. The fundamental commitment of this paper is a strategy for the early location and separation of sugar beet infections in view of Support Vector Machines and phantom vegetation files. The point was to separate sick from non-ailing sugar beet leaves, to separate between the sicknesses Cercospora leaf spot, leaf rust and fine mold, and to recognize infections even before unambiguous side effects became apparent.
