The present invention relates to a method, apparatus and a
system of fast diagnosis of stresses and diseases in higher plants. The proposed methodology is based on the
hypothesis of that when a
plant is in imbalance; there are changes in its
metabolism that render an alteration of the
chemical composition of its organs. This chemical alteration leads to a change in the physical properties, such as the
fluorescence of the leaves. Due to the complexity of the material of the leaves, the
present method proposes that the
signal be treated with statistical methods and that the classification is made through softwares based on
machine learning. As an example of the application of the invention, the results are shown for the
Greening disease in citrus. Currently,
Greening is the most severe citrus
disease since there is no treatment available for it and due to its high
dissemination rate and the fact that it affects all varieties of orange trees, being the diagnosis performed through
visual inspection, which renders high subjectivity, high error percentage and the
disease is only diagnosed after the expression of the symptoms (˜8 months). During the
asymptomatic phase, the infected tree is a source of
dissemination of the disease. The present invention can perform the
asymptomatic diagnosis of
Greening disease from the leaf with a percentage of correct diagnosis higher than 80%.